Modern Control Systems

**3**

**Chapter 1**

**Abstract**

train control system

**1. Introduction**

Study on a New Train Control

Viewpoint of Safety2.0

*Hideo Nakamura and Yoshihisa Saito*

System in the IoT Era: From the

Safety2.0 which advocates cooperative safety is attracting attention. Assuming that Industry4.0 proposed by the German authorities is an IoT-based production revolution, Safety2.0 is a Japanese-originated proposal that seeks to create a more flexible and sophisticated safety by introducing Internet of Things (IoT) into production sites. This chapter introduces the concepts of Safety2.0 and its spread internationally, focusing on the activities of IGSAP, a Safety2.0 promoter. Furthermore, we look back on the conventional train control from the viewpoint of Safetyx.x and look at the appearance of the train control suitable for Safety2.0 using IoT. As a result, in this chapter, we propose a simple and smart train control system unified train control system (UTCS), in which a train control system is realized in a hierarchical structure of a logic layer, a

The safety of manufacturing today is mainly based on European-style methodology mainly based on certification found in functional safety standards (IEC61508) and the like. Since stakeholders have different standing positions for certification, there is a reflection that productivity cannot be improved by mutually balancing each other. Safety2.0 focuses on the realization of "cooperative safety" that creates a highly safe condition by mutually exchanging information among the elements that constitute systems based on IoT. Therefore, under Safety2.0, stakeholders are required to share their wisdom with each other and to manufacture products based on IoT. What is the meaning of "relying on IoT" for railways? Radio train control systems advanced train administration and communications system (ATACS) and communications-based train control (CBTC) will be analyzed and evaluated from

network layer, and a terminal layer, and discuss its processing method.

the viewpoint of Safety2.0, and future train control will be considered.

The new safety concept named Safety2.0 is not unrelated to the desire to overcome occupational safety occlusions at production sites. In the summer of 2015, the Safety2.0 Preparatory Committee (renamed the "Safety2.0 Promotion Committee"

**2. Concept and current state of Safety2. 0**

**Keywords:** Industry4.0, IoT, Safety2.0, UTCS, ATP-block system,

#### **Chapter 1**

## Study on a New Train Control System in the IoT Era: From the Viewpoint of Safety2.0

*Hideo Nakamura and Yoshihisa Saito*

#### **Abstract**

Safety2.0 which advocates cooperative safety is attracting attention. Assuming that Industry4.0 proposed by the German authorities is an IoT-based production revolution, Safety2.0 is a Japanese-originated proposal that seeks to create a more flexible and sophisticated safety by introducing Internet of Things (IoT) into production sites. This chapter introduces the concepts of Safety2.0 and its spread internationally, focusing on the activities of IGSAP, a Safety2.0 promoter. Furthermore, we look back on the conventional train control from the viewpoint of Safetyx.x and look at the appearance of the train control suitable for Safety2.0 using IoT. As a result, in this chapter, we propose a simple and smart train control system unified train control system (UTCS), in which a train control system is realized in a hierarchical structure of a logic layer, a network layer, and a terminal layer, and discuss its processing method.

**Keywords:** Industry4.0, IoT, Safety2.0, UTCS, ATP-block system, train control system

#### **1. Introduction**

The safety of manufacturing today is mainly based on European-style methodology mainly based on certification found in functional safety standards (IEC61508) and the like. Since stakeholders have different standing positions for certification, there is a reflection that productivity cannot be improved by mutually balancing each other. Safety2.0 focuses on the realization of "cooperative safety" that creates a highly safe condition by mutually exchanging information among the elements that constitute systems based on IoT. Therefore, under Safety2.0, stakeholders are required to share their wisdom with each other and to manufacture products based on IoT. What is the meaning of "relying on IoT" for railways? Radio train control systems advanced train administration and communications system (ATACS) and communications-based train control (CBTC) will be analyzed and evaluated from the viewpoint of Safety2.0, and future train control will be considered.

#### **2. Concept and current state of Safety2. 0**

The new safety concept named Safety2.0 is not unrelated to the desire to overcome occupational safety occlusions at production sites. In the summer of 2015, the Safety2.0 Preparatory Committee (renamed the "Safety2.0 Promotion Committee" in 2016) was established. The conclusion was that the essential elements of the system exchanged information with each other to create optimal safety, which was the construction of a cooperative safety methodology suitable for the IoT era [1, 2]. Safety2.0 was, of course, preceded by Safety0.0 and Safety1.0, which supported present-day safety. However, the use of IoT is very effective in overcoming the sense of occlusion on it. In this chapter, we review the changes in safety initiatives from the perspective of Safety0.0 and Safety1.0, and confirm the today's status of Safety2.0. In addition, practical activities aimed at realizing Safety2.0 have begun, and we would like to introduce the situation.

#### **2.1 Safety0.0 to maintain safety with sustained arousal**

Direction calls are famous for railway safety culture. This is also an easy-tounderstand case of Safety0.0, which attempts to prevent accidents by drawing attention and keeping the spirit awake at all times. Japan has been regarded as a leader in the sustainable implementation of the "Zero-Accident Motion" and other activities leading to Safety0.0. Looking back at this, various activities have been carried out, including the enactment of the Occupational Safety and Health Law in 1972 and the start of the "Everyone's Participation in Zero Accidents" motion by the Center Industrial Accident Prevention Association in 1973. On the other hand, it is interesting to note that the European version of the Zero-Disaster Motion, which is a top-down movement but corresponds to Safety0.0, is beginning to take place in Europe, where the International Society for Social Security (ISSA; International Social Security Association) has achieved outcomes at Safety1.0 through the launch of Vision Zero and the launch of Zero Accidents Forum by Europe and Finland.

#### **2.2 Safety1.0 to prevent accidents**

There are limitations to ensuring safety by Safety0.0 alone, since no mistakes will be made and the machinery will be destroyed, no matter how well people are trained. Therefore, technological efforts to implement some sort of safety measures for "goods" such as machinery and systems have progressed on the assumption that "people make mistakes" and "machinery breaks down." In Europe, this was established as a mandatory standard, and work accidents were greatly reduced by providing industrial machinery with safety protection measures. The basic idea was to establish a barrier between industrial machinery, which is a hazardous source, and humans, and to establish a mechanism for moving machinery only when humans are absent. This is Safety1.0. In Japan, the Industrial Safety and Health Law was revised in 2005, and risk assessment was added as an obligation to make initiatives. Efforts learned from European experiences and outcomes are now being developed. For this implementation, a number of safety mechanisms have been developed and incorporated into devices at each industrial site. However, even though it is isolated from hazardous sources, it is not possible to completely isolate workers during maintenance. In addition, many sites are difficult to isolate, such as construction work.

#### **2.3 IoT-based Safety2.0 concepts**

Safety2.0 is a system that ensures methodology suitable for the IoT era, in which the essential elements constituting the system exchange information with each other to achieve optimal safety [1, 2]. In this respect, it is different from the idea of relying on human attention (Safety0.0) and of taking some protective measures against human errors and mechanical failures to ensure safety (Safety1.0). These relationships are summarized in **Figure 1**. In Safety0.0, risks exist in a wide range, including

**5**

**Figure 1.**

*Comparison among Safety(x.x).*

Promotion) was established.

*Study on a New Train Control System in the IoT Era: From the Viewpoint of Safety2.0*

coexistence areas, in order to prevent accidents due to attention and judgment. On the other hand, in Safety1.0, the risks were reduced by dividing the human area into the machinery domain and by providing various safety measures in the machine domain.

On the other hand, in Safety2.0, machineries and humans exchange information and cooperate with each other to ensure safety, so that both machines and humans can coexist with each other. In addition, since appropriate safety using information

The Nikkei BP brochure [1] explains Safety2.0 as follows: "Frankly speaking Safety2.0 is a collaborative safety built by people, goods, and the environmental in cooperation with each other. The best mean to achieve this Safety2.0 is to make rapid progress today. In Safety1.0, there was only a choice between "stop" and "go." In Safety2.0, however, detailed operations are carried out by exchanging information between people and machineries, and the safely coexistence between both is aimed at (Omitted hereafter.)." A case of Safety2.0 is indicated in **Figures 2** and **3**. Management understandings and support are essential to Safety2.0's in-house promotion. Fortunately, IoT-based safety technologies were widely developed prior to the conceptual development of Safety2.0, and there were certain grounds to accept Safety2.0. However, in order to expand Safety2.0 from Japan to a wide range of countries as well as in Japan, we would like to have a promotion base. To this end, the Safety Global Promotion Mechanism (IGSAP; The Institute of Global Safety

IGSAP has established the safety management forum as a forum for managers and managers to gather and replace information and experiences in order to actively

Safety of isolation that creates as little coexistence as possible is fundamental.

is maintained during operation, the overall risk is greatly reduced.

**2.4 Current stage of Safety2.0 (activities centered on IGSAP)**

*DOI: http://dx.doi.org/10.5772/intechopen.80306*

*Study on a New Train Control System in the IoT Era: From the Viewpoint of Safety2.0 DOI: http://dx.doi.org/10.5772/intechopen.80306*

#### **Figure 1.**

*Applied Modern Control*

and we would like to introduce the situation.

**2.2 Safety1.0 to prevent accidents**

**2.3 IoT-based Safety2.0 concepts**

**2.1 Safety0.0 to maintain safety with sustained arousal**

in 2016) was established. The conclusion was that the essential elements of the system exchanged information with each other to create optimal safety, which was the construction of a cooperative safety methodology suitable for the IoT era [1, 2]. Safety2.0 was, of course, preceded by Safety0.0 and Safety1.0, which supported present-day safety. However, the use of IoT is very effective in overcoming the sense of occlusion on it. In this chapter, we review the changes in safety initiatives from the perspective of Safety0.0 and Safety1.0, and confirm the today's status of Safety2.0. In addition, practical activities aimed at realizing Safety2.0 have begun,

Direction calls are famous for railway safety culture. This is also an easy-tounderstand case of Safety0.0, which attempts to prevent accidents by drawing attention and keeping the spirit awake at all times. Japan has been regarded as a leader in the sustainable implementation of the "Zero-Accident Motion" and other activities leading to Safety0.0. Looking back at this, various activities have been carried out, including the enactment of the Occupational Safety and Health Law in 1972 and the start of the "Everyone's Participation in Zero Accidents" motion by the Center Industrial Accident Prevention Association in 1973. On the other hand, it is interesting to note that the European version of the Zero-Disaster Motion, which is a top-down movement but corresponds to Safety0.0, is beginning to take place in Europe, where the International Society for Social Security (ISSA; International Social Security Association) has achieved outcomes at Safety1.0 through the launch of Vision Zero and the launch of Zero Accidents Forum by Europe and Finland.

There are limitations to ensuring safety by Safety0.0 alone, since no mistakes will be made and the machinery will be destroyed, no matter how well people are trained. Therefore, technological efforts to implement some sort of safety measures for "goods" such as machinery and systems have progressed on the assumption that "people make mistakes" and "machinery breaks down." In Europe, this was established as a mandatory standard, and work accidents were greatly reduced by providing industrial machinery with safety protection measures. The basic idea was to establish a barrier between industrial machinery, which is a hazardous source, and humans, and to establish a mechanism for moving machinery only when humans are absent. This is Safety1.0. In Japan, the Industrial Safety and Health Law was revised in 2005, and risk assessment was added as an obligation to make initiatives. Efforts learned from European experiences and outcomes are now being developed. For this implementation, a number of safety mechanisms have been developed and incorporated into devices at each industrial site. However, even though it is isolated from hazardous sources, it is not possible to completely isolate workers during maintenance. In addition, many sites are difficult to isolate, such as construction work.

Safety2.0 is a system that ensures methodology suitable for the IoT era, in which the essential elements constituting the system exchange information with each other to achieve optimal safety [1, 2]. In this respect, it is different from the idea of relying on human attention (Safety0.0) and of taking some protective measures against human errors and mechanical failures to ensure safety (Safety1.0). These relationships are summarized in **Figure 1**. In Safety0.0, risks exist in a wide range, including

**4**

*Comparison among Safety(x.x).*

coexistence areas, in order to prevent accidents due to attention and judgment. On the other hand, in Safety1.0, the risks were reduced by dividing the human area into the machinery domain and by providing various safety measures in the machine domain. Safety of isolation that creates as little coexistence as possible is fundamental.

On the other hand, in Safety2.0, machineries and humans exchange information and cooperate with each other to ensure safety, so that both machines and humans can coexist with each other. In addition, since appropriate safety using information is maintained during operation, the overall risk is greatly reduced.

The Nikkei BP brochure [1] explains Safety2.0 as follows: "Frankly speaking Safety2.0 is a collaborative safety built by people, goods, and the environmental in cooperation with each other. The best mean to achieve this Safety2.0 is to make rapid progress today. In Safety1.0, there was only a choice between "stop" and "go." In Safety2.0, however, detailed operations are carried out by exchanging information between people and machineries, and the safely coexistence between both is aimed at (Omitted hereafter.)." A case of Safety2.0 is indicated in **Figures 2** and **3**.

Management understandings and support are essential to Safety2.0's in-house promotion. Fortunately, IoT-based safety technologies were widely developed prior to the conceptual development of Safety2.0, and there were certain grounds to accept Safety2.0. However, in order to expand Safety2.0 from Japan to a wide range of countries as well as in Japan, we would like to have a promotion base. To this end, the Safety Global Promotion Mechanism (IGSAP; The Institute of Global Safety Promotion) was established.

#### **2.4 Current stage of Safety2.0 (activities centered on IGSAP)**

IGSAP has established the safety management forum as a forum for managers and managers to gather and replace information and experiences in order to actively engage in the safety of customers, employees, and the safety of the company as a company. In addition, Japan has been vigorously inviting European opinion leaders to work to eradicate occupational accidents under the Vision Zero and replace opinions through visits to Europe. In February 2018, NIPPO's automatic stop tire roller/wheel loader, equipped with an emergency stop technology for construction machinery, was the first Safety2.0 automatic stop tire roller/wheel loader to be registered.

**7**

*Study on a New Train Control System in the IoT Era: From the Viewpoint of Safety2.0*

**3. Outlook for train control systems from the viewpoint of Safety2.0**

Following the opening of the railway, the railway construction regulations and railway dormitory train transport regulations were enacted, and other rules for ensuring safety operation were rapidly established. This is a summary of the standards and regulations that form the base of Safety0.0. Technologies have also been imported overseas. For example, 1887, a voucher-type blocking method with the use of occlusion telegraphs was established between Kyoto and Osaka, and a secondclass mechanical interlocking device was installed at the junction of the Yamanote Line and the Tokaido Line in Shinagawa Station. In addition, the manufacture of railway signals and interlocking equipment began at the Mimura Factory of Tokyo

Tsukishima, and the move toward in-house production also progressed [3].

attention in the meaning that mistakes could lead to accidents as they were.

confirming safety are fail-safe, and the Safety1.0 mode is constructed.

Signaling devices based on these technologies are mainly designed to ensure routes and link signals, and the safety of drivers can be guaranteed if they operate according to the signals. However, there was an accident caused by a mistake by the station manager that the train line between stations was forgotten and the opposite route was set in the single-line section. The emphasis was placed on raising human

Machinery is a hazard source in the world of machine safety, such as factories. For this reason, a mechanism for confirming that no human or body part exists in the work area of the machinery and allowing the machine to operate only at that time has been adopted. Furthermore, the sensors used therein and the devices for

On the other hand, trains are the main source of hazard in railways. For this reason, the "concept of isolation" that allows only one train to exist in one section was established as a blockage. In addition, in the station premises, an interlocking device has been developed, which ensures absolute safety even if erroneous signal handling is performed. In addition, the use of fail-safe orbit circuits has resulted in advanced signal systems. However, even under these mechanisms, mistakes of the driver, such as signal advancement, can cause an accident. For this reason, in-vehicle alarms have been developed, which give an alarm to wake up when a stop signal is approached, and they have evolved to an ATS, which applies an emergency brake when a driver does not perform a

However, the technology introduced in the field of railways and industrial machinery that is safe but that is basically safe to stop was not the same as the technology introduced in the case where flight continuity is safe instead of stopping as in the case of an aircraft. In addition, own technologies were developed in the industrial machinery field and railways. As a result, no agreement was reached on common safety and fail-safe technologies across industries, and there was no com-

In the 1980s, computers were introduced into Safety1.0, which had been secured with sophisticated circuitry. As a result, the aspects of safety technologies that have been uniquely pursued in each industry have changed. In addition to the conventional technology, the concern of engineers in various industries has been to ensure

On October 14, 1872, Japan's railway opened between Shimbashi and Yokohama.

*DOI: http://dx.doi.org/10.5772/intechopen.80306*

**3.1 Safety1.0 and train control**

predetermined treatment.

mon measure of safety.

*3.1.1 Actual state of safety assurance based on Safety1.0*

*3.1.2 Computer-based safety control device and Safety1.0*

#### **Figure 2.** *An example of Safety2.0 (Case 1).*

**Figure 3.** *An example of Safety2.0 (Case 2).*

#### **3. Outlook for train control systems from the viewpoint of Safety2.0**

On October 14, 1872, Japan's railway opened between Shimbashi and Yokohama. Following the opening of the railway, the railway construction regulations and railway dormitory train transport regulations were enacted, and other rules for ensuring safety operation were rapidly established. This is a summary of the standards and regulations that form the base of Safety0.0. Technologies have also been imported overseas. For example, 1887, a voucher-type blocking method with the use of occlusion telegraphs was established between Kyoto and Osaka, and a secondclass mechanical interlocking device was installed at the junction of the Yamanote Line and the Tokaido Line in Shinagawa Station. In addition, the manufacture of railway signals and interlocking equipment began at the Mimura Factory of Tokyo Tsukishima, and the move toward in-house production also progressed [3].

Signaling devices based on these technologies are mainly designed to ensure routes and link signals, and the safety of drivers can be guaranteed if they operate according to the signals. However, there was an accident caused by a mistake by the station manager that the train line between stations was forgotten and the opposite route was set in the single-line section. The emphasis was placed on raising human attention in the meaning that mistakes could lead to accidents as they were.

#### **3.1 Safety1.0 and train control**

*Applied Modern Control*

engage in the safety of customers, employees, and the safety of the company as a company. In addition, Japan has been vigorously inviting European opinion leaders to work to eradicate occupational accidents under the Vision Zero and replace opinions through visits to Europe. In February 2018, NIPPO's automatic stop tire roller/wheel loader, equipped with an emergency stop technology for construction machinery, was the first Safety2.0 automatic stop tire roller/wheel loader to be registered.

**6**

**Figure 3.**

**Figure 2.**

*An example of Safety2.0 (Case 1).*

*An example of Safety2.0 (Case 2).*

#### *3.1.1 Actual state of safety assurance based on Safety1.0*

Machinery is a hazard source in the world of machine safety, such as factories. For this reason, a mechanism for confirming that no human or body part exists in the work area of the machinery and allowing the machine to operate only at that time has been adopted. Furthermore, the sensors used therein and the devices for confirming safety are fail-safe, and the Safety1.0 mode is constructed.

On the other hand, trains are the main source of hazard in railways. For this reason, the "concept of isolation" that allows only one train to exist in one section was established as a blockage. In addition, in the station premises, an interlocking device has been developed, which ensures absolute safety even if erroneous signal handling is performed. In addition, the use of fail-safe orbit circuits has resulted in advanced signal systems. However, even under these mechanisms, mistakes of the driver, such as signal advancement, can cause an accident. For this reason, in-vehicle alarms have been developed, which give an alarm to wake up when a stop signal is approached, and they have evolved to an ATS, which applies an emergency brake when a driver does not perform a predetermined treatment.

However, the technology introduced in the field of railways and industrial machinery that is safe but that is basically safe to stop was not the same as the technology introduced in the case where flight continuity is safe instead of stopping as in the case of an aircraft. In addition, own technologies were developed in the industrial machinery field and railways. As a result, no agreement was reached on common safety and fail-safe technologies across industries, and there was no common measure of safety.

#### *3.1.2 Computer-based safety control device and Safety1.0*

In the 1980s, computers were introduced into Safety1.0, which had been secured with sophisticated circuitry. As a result, the aspects of safety technologies that have been uniquely pursued in each industry have changed. In addition to the conventional technology, the concern of engineers in various industries has been to ensure

the safety of the computer itself (hardware) used and to prevent bugs (quality assurance) in the software to be incorporated.

Computer hardware safety has been solved by redundant configuration and verification of processing results and appropriate integration of diagnostic circuitry, but the methodology has been discussed across industries. Software has a common issue: how to develop high-quality, bug-free software. What is important is that the sophisticated circuits that have once been inherited as the essence of safety technology by various industry sectors have all been incorporated as software logic and have not appeared on the surface.

This resulted in a deeper recognition of common methodologies across industrials and the establishment of a new IEC61508 of international standards that can be applied across industries under the new concept of "functional safety." In order to ensure safety in the age of functional safety, first of all, the safety level of the target systems is determined as safety integrity levels (SILs) as a result of the risk analysis. In addition, the design requirements of the hardware and the targets of the hazardous side failure rate are indicated in accordance with the SIL value. In software, the design requirements for each phase of the life cycle are determined according to the SIL value. Thus, "risk" became a common measure of safety. On the other hand, the authenticating organization has evaluated the validation of the determination of the SIL value and the validity of the specific work according to the value of the SIL.

The situation is different in the age of reliance on circuit technology and in the age of use of computers. Nevertheless, the concept of Safety1.0, which seeks to safeguard safety by protection measures in the event of human error or device defect, is common.

What is important is the fact that computer use is evolving the control system into a more sophisticated one. There is a great difference between the age and today of the development of computerized signaling devices that have solved the issue of how to make hardware and software safe and have replaced safety technology with program logic to produce electronic interlocks and level crossings.

The elimination of concerns about the use of computers for safety control has facilitated the addition of advanced functions. In addition, the successful use of diagnostic technology has opened the way to integrate communication technologies such as networks and wirelesses into safety control devices. The issue of "train control system using IoT" is also due to the fact that communication including wireless communication can be freely used for safety control. Today, the challenge of developing new computer-based systems is continuing, and sophisticated control systems are emerging. In 2011, the world-first full-fledged wireless-train control systems ATACS was launched on the Sensei Line and has been achieving excellent results. Based on this achievement, it was also introduced between Ikebukuro and Omiya on the Saikyo Line in 2016, and has been operating stably. The SPARCS (simple-structure and high-performance ATC by wireless communication system) of radio train control systems developed by Nippon Signal Co. Ltd. is also well received oversea. What are the relationships between these advanced systems and Safety2.0? What kind of system should we look at next to these advanced systems? We would like to consider on the base of the specifications of the computer- and radio-aided train (CARAT) control system in which the author participated in the development when he was in the Railway Research Institute.

#### **3.2 Advanced train control systems and Safety2.0**

The interlocking of the CARAT is called point-control. The position of the train is managed by the block-ID and the position (kilometer) in the block. A plurality of trains can exist in one block except for the section of the point machine in order to realize the movement blocking even in the station premises. Point control does not include path locking or segmented locking. Since the route is pulled back by

**9**

**Figure 4.**

*An architecture of conventional train control system.*

*Study on a New Train Control System in the IoT Era: From the Viewpoint of Safety2.0*

exchanging information with the on-board device, it is not determined only by the train position. A reasonable and safe process is substituted for the time of the access lock, which was uniformly applied when the line is in the approaching section. The prototype of this point control was installed at Tsubame-Sanjō Station, and the

Since the CARAT was designed to cope with the Shinkansen, the level crossing control function is not required. However, investigation of functions and check of the effect by simulation were carried out, and it was proven to be effective for the fixed-time control and safety improvement, which had been regarded as an issue of the existing level crossing. As a result, the point control and the level crossing control are positioned as processes for extending the point where the train is allowed to travel. The interval control device generates the "traveling permit point information" and transmits the result to the on-board device, thereby making it possible to unify the processing in the interim of the station and in the premises of the station. The form in which each of the emerging devices performs reasonable processing while exchanging information in various directional is precisely located in the IoT. The compatibility between CBTC and Safety2.0 appears to be good.

Existing train control systems have condensed know-how learned from the experience of large accidents caused by human error. As shown in **Figure 4**, the basic control function is the blocking function and the interlocking function. However, safety cannot be ensured by this function alone. Today's safety is achieved in cooperation with safety devices such as ATS and ATC for the objective of preventing accidents caused by human error. Nevertheless, as shown in **Figure 4**, the

A simple and orderly system as shown in **Figure 5** emerges from the IoT-based train control system. On the scene, there are only point machines, level crossings, and trains that make up the route of travel. The processing unit of the center directly transmits "information up to the section where the vehicle can safely travel" to the on-board device of the train as a control command (travel command). In the CARAT, the interval control device sent the "travel permission point information" to the company office device below the information of the point control. The interval control device is centralized at the center, and at the same time, both the point control and the level crossing control are centralized at the center. Therefore, this form is organized only by adding the IoT viewpoint to that demonstrated in the CARAT.

*DOI: http://dx.doi.org/10.5772/intechopen.80306*

function was confirmed by a monitor run.

**3.3 Outlook for future train control systems**

actual situation is a complex combination.

*Study on a New Train Control System in the IoT Era: From the Viewpoint of Safety2.0 DOI: http://dx.doi.org/10.5772/intechopen.80306*

exchanging information with the on-board device, it is not determined only by the train position. A reasonable and safe process is substituted for the time of the access lock, which was uniformly applied when the line is in the approaching section. The prototype of this point control was installed at Tsubame-Sanjō Station, and the function was confirmed by a monitor run.

Since the CARAT was designed to cope with the Shinkansen, the level crossing control function is not required. However, investigation of functions and check of the effect by simulation were carried out, and it was proven to be effective for the fixed-time control and safety improvement, which had been regarded as an issue of the existing level crossing. As a result, the point control and the level crossing control are positioned as processes for extending the point where the train is allowed to travel. The interval control device generates the "traveling permit point information" and transmits the result to the on-board device, thereby making it possible to unify the processing in the interim of the station and in the premises of the station. The form in which each of the emerging devices performs reasonable processing while exchanging information in various directional is precisely located in the IoT. The compatibility between CBTC and Safety2.0 appears to be good.

#### **3.3 Outlook for future train control systems**

Existing train control systems have condensed know-how learned from the experience of large accidents caused by human error. As shown in **Figure 4**, the basic control function is the blocking function and the interlocking function. However, safety cannot be ensured by this function alone. Today's safety is achieved in cooperation with safety devices such as ATS and ATC for the objective of preventing accidents caused by human error. Nevertheless, as shown in **Figure 4**, the actual situation is a complex combination.

A simple and orderly system as shown in **Figure 5** emerges from the IoT-based train control system. On the scene, there are only point machines, level crossings, and trains that make up the route of travel. The processing unit of the center directly transmits "information up to the section where the vehicle can safely travel" to the on-board device of the train as a control command (travel command). In the CARAT, the interval control device sent the "travel permission point information" to the company office device below the information of the point control. The interval control device is centralized at the center, and at the same time, both the point control and the level crossing control are centralized at the center. Therefore, this form is organized only by adding the IoT viewpoint to that demonstrated in the CARAT.

**Figure 4.**

*An architecture of conventional train control system.*

*Applied Modern Control*

assurance) in the software to be incorporated.

and have not appeared on the surface.

the safety of the computer itself (hardware) used and to prevent bugs (quality

Computer hardware safety has been solved by redundant configuration and verification of processing results and appropriate integration of diagnostic circuitry, but the methodology has been discussed across industries. Software has a common issue: how to develop high-quality, bug-free software. What is important is that the sophisticated circuits that have once been inherited as the essence of safety technology by various industry sectors have all been incorporated as software logic

This resulted in a deeper recognition of common methodologies across industrials and the establishment of a new IEC61508 of international standards that can be applied across industries under the new concept of "functional safety." In order to ensure safety in the age of functional safety, first of all, the safety level of the target systems is determined as safety integrity levels (SILs) as a result of the risk analysis. In addition, the design requirements of the hardware and the targets of the hazardous side failure rate are indicated in accordance with the SIL value. In software, the design requirements for each phase of the life cycle are determined according to the SIL value. Thus, "risk" became a common measure of safety. On the other hand, the authenticating organization has evaluated the validation of the determination of the SIL value

The situation is different in the age of reliance on circuit technology and in the age of use of computers. Nevertheless, the concept of Safety1.0, which seeks to safeguard safety by protection measures in the event of human error or device defect, is common. What is important is the fact that computer use is evolving the control system into a more sophisticated one. There is a great difference between the age and today of the development of computerized signaling devices that have solved the issue of how to make hardware and software safe and have replaced safety technology with

The elimination of concerns about the use of computers for safety control has facilitated the addition of advanced functions. In addition, the successful use of diagnostic technology has opened the way to integrate communication technologies such as networks and wirelesses into safety control devices. The issue of "train control system using IoT" is also due to the fact that communication including wireless communication can be freely used for safety control. Today, the challenge of developing new computer-based systems is continuing, and sophisticated control systems are emerging. In 2011, the world-first full-fledged wireless-train control systems ATACS was launched on the Sensei Line and has been achieving excellent results. Based on this achievement, it was also introduced between Ikebukuro and Omiya on the Saikyo Line in 2016, and has been operating stably. The SPARCS (simple-structure and high-performance ATC by wireless communication system) of radio train control systems developed by Nippon Signal Co. Ltd. is also well received oversea. What are the relationships between these advanced systems and Safety2.0? What kind of system should we look at next to these advanced systems? We would like to consider on the base of the specifications of the computer- and radio-aided train (CARAT) control system in which the author participated in the

The interlocking of the CARAT is called point-control. The position of the train is managed by the block-ID and the position (kilometer) in the block. A plurality of trains can exist in one block except for the section of the point machine in order to realize the movement blocking even in the station premises. Point control does not include path locking or segmented locking. Since the route is pulled back by

and the validity of the specific work according to the value of the SIL.

program logic to produce electronic interlocks and level crossings.

development when he was in the Railway Research Institute.

**3.2 Advanced train control systems and Safety2.0**

**8**

**Figure 5.** *System architecture of unified train control system.*

The "interlocking device," "blocking device," and "ATC/ATS," which have been so popular in the signal field as to be the three types of gods, disappear. However, it is not the introduction of the "centralized linkage device." Existing interlocks have incorporated various locking logics to ensure the safety of whatever handling is done by the signal handler. In this respect, the centralized interlocking device does not change at all. Instead, it is claiming to replace the complex interlocking logic itself with point control, which makes it unnecessary. Point-control algorithms have been demonstrated in the monitoring run of the next-generation train control systems CARAT carried out by the Railway Institute on the Joetsu Shinkansen.

The overall system architecture consists of a terminal layer and a center device (functional layer) for controlling site equipment/on-board safety control devices, and an IP network (network layer) connecting them. This level next-generation system is named unified train control system (UTCS), and various studies are being conducted [5, 6]. This system also conforms to the concept of Safety2.0, which consists essentially only of the equipment necessary for the system: trains, point machines, level crossings, and center equipment, and "the essentially necessary equipment exchanges information with each other and realizes functions (I have called this intrinsic control)."

#### **3.4 Outline of the UTCS processes**

In the UTCS, the concept of a "path" (labeled "authorized route" in **Figure 5**) for a train is introduced for the standardization of processes. A path means a "limit position to which running is possible," and is derived from an associated preceding train, a point machine, and the states of a level crossing for each train. For this reason, train processes by unified processors are realized by train tracking, path searching (or "route searching"), and control processes that are initiated by route searching processes in order to control level crossings and point machines.

When paths for trains are determined, an authorized command with additional speed restriction information in a path is also generated and sent to the corresponding terminal device of the terminal layer. Path searching creates a search for a limit point to which running is possible (a path) in the train movement direction. In

**11**

**Figure 7.**

*Components of ATP-block system.*

**Figure 6.**

*Data flow under UTCS.*

*Study on a New Train Control System in the IoT Era: From the Viewpoint of Safety2.0*

the case of station premises, however, a search is made according to a scheduled running path acquired from the running control device (or "traffic control system") on the functional layer. The path at this time is based on the terminal end of the running path and is determined by the state of point machines existing in between and at the tail position of a possible preceding train (including the safety margin). On the other hand, in the case of a midway point between stations, the tail position of the preceding train or the state (labeled "status" in **Figure 5**) of an existing level crossing is associated with the determination of a path. If the level crossing is controlled by the relevant train and the status indicates "passing allowed," which means closing completion and no obstacle, the search is extended up to a further remote position. Although on-board devices are responsible for on-board safety processing, a continuous speed check according to a pattern is realized on the train anyway. Moreover, in the case of the CBTC, a high-level speed check function can be realized by installing a terminal device on the train, rather than providing an ATP terminal device on the ground. In **Figure 6**, a data flow under UTCS is illustrated. An example of a UTCS that relies on Safety2.0 is ATP-block system [4–6], which is typical of intrinsic control, although the detail of the ATP-block system is omitted, and

*DOI: http://dx.doi.org/10.5772/intechopen.80306*

*Study on a New Train Control System in the IoT Era: From the Viewpoint of Safety2.0 DOI: http://dx.doi.org/10.5772/intechopen.80306*

the case of station premises, however, a search is made according to a scheduled running path acquired from the running control device (or "traffic control system") on the functional layer. The path at this time is based on the terminal end of the running path and is determined by the state of point machines existing in between and at the tail position of a possible preceding train (including the safety margin).

On the other hand, in the case of a midway point between stations, the tail position of the preceding train or the state (labeled "status" in **Figure 5**) of an existing level crossing is associated with the determination of a path. If the level crossing is controlled by the relevant train and the status indicates "passing allowed," which means closing completion and no obstacle, the search is extended up to a further remote position. Although on-board devices are responsible for on-board safety processing, a continuous speed check according to a pattern is realized on the train anyway. Moreover, in the case of the CBTC, a high-level speed check function can be realized by installing a terminal device on the train, rather than providing an ATP terminal device on the ground. In **Figure 6**, a data flow under UTCS is illustrated.

An example of a UTCS that relies on Safety2.0 is ATP-block system [4–6], which is typical of intrinsic control, although the detail of the ATP-block system is omitted, and

**Figure 6.** *Data flow under UTCS.*

*Applied Modern Control*

**Figure 5.**

The "interlocking device," "blocking device," and "ATC/ATS," which have been so popular in the signal field as to be the three types of gods, disappear. However, it is not the introduction of the "centralized linkage device." Existing interlocks have incorporated various locking logics to ensure the safety of whatever handling is done by the signal handler. In this respect, the centralized interlocking device does not change at all. Instead, it is claiming to replace the complex interlocking logic itself with point control, which makes it unnecessary. Point-control algorithms have been demonstrated in the monitoring run of the next-generation train control systems CARAT carried out by the Railway Institute on the Joetsu Shinkansen. The overall system architecture consists of a terminal layer and a center device (functional layer) for controlling site equipment/on-board safety control devices, and an IP network (network layer) connecting them. This level next-generation system is named unified train control system (UTCS), and various studies are being conducted [5, 6]. This system also conforms to the concept of Safety2.0, which consists essentially only of the equipment necessary for the system: trains, point machines, level crossings, and center equipment, and "the essentially necessary equipment exchanges information with each other and realizes functions (I have

In the UTCS, the concept of a "path" (labeled "authorized route" in **Figure 5**) for a train is introduced for the standardization of processes. A path means a "limit position to which running is possible," and is derived from an associated preceding train, a point machine, and the states of a level crossing for each train. For this reason, train processes by unified processors are realized by train tracking, path searching (or "route searching"), and control processes that are initiated by route

When paths for trains are determined, an authorized command with additional speed restriction information in a path is also generated and sent to the corresponding terminal device of the terminal layer. Path searching creates a search for a limit point to which running is possible (a path) in the train movement direction. In

searching processes in order to control level crossings and point machines.

**10**

called this intrinsic control)."

**3.4 Outline of the UTCS processes**

*System architecture of unified train control system.*

**Figure 7.** *Components of ATP-block system.*

the essential devices constituting the system mutually exchange information as IoTs, thereby realizing advanced functions. In the case of ATP-block system, the blocking device and the interlocking device, which were previously located at the station and controlling the driving direction with the adjacent station, disappear(see **Figures 7**, **8**) The history of technological progress in a train control system and its relation to the Safety(x,x) is illustrated by **Figure 9**.

#### **Figure 8.**

*System configuration of ATP-block system.*


**Figure 9.** *History of technological progress in a train control system and Safety(x.x).*

#### **4. Afterword and conclusion**

Development of new systems involves certification work in accordance with international standards. Especially in the train control, the train is subjected to the baptism of the standard of the reliability, availability, maintainability and safety (RAMS; IEC62278). To make this baptization smart, it avoids complications and makes the system as simple as possible. Railways are one of the social systems and have a long service life. Erroneous selects can leave the roots of the trouble. Examining and simplifying the components as much as possible improves the system's visibility and facilitates certification. Furthermore, the reliability is increased by the reduction of the amount of goods, and protection becomes unnecessary. The

**13**

**Author details**

Hideo Nakamura1

provided the original work is properly cited.

\* and Yoshihisa Saito2

2 Kyosan Electric Mfg. Co., Ltd., Yokohama-Shi, Kanagawa, Japan

\*Address all correspondence to: nakamura.hideo@nihon-u.ac.jp

1 Visiting Researcher of the University of Tokyo, Japan

© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium,

*Study on a New Train Control System in the IoT Era: From the Viewpoint of Safety2.0*

fewer the number of interfaces, the greater the safety. In addition to the advantages of these nonfunctional requirements, we believe that systematization in accordance with Safety2.0 can be an informative methodology in this regard. Under these circumstances, if information such as orbit protection is automatically extracted by AI based on vehicle vibration data and the like during daily driving, the railway can

Safety2.0 is an initiative that contributes not only to safety effects but also to productivity improvements and contributes to management. I hope that UTCS will be a successful development. We believe that this will also contribute to the dissemination

*DOI: http://dx.doi.org/10.5772/intechopen.80306*

of Japanese Safety2.0.

be reconstructed as a competitive transportation means.

#### *Study on a New Train Control System in the IoT Era: From the Viewpoint of Safety2.0 DOI: http://dx.doi.org/10.5772/intechopen.80306*

fewer the number of interfaces, the greater the safety. In addition to the advantages of these nonfunctional requirements, we believe that systematization in accordance with Safety2.0 can be an informative methodology in this regard. Under these circumstances, if information such as orbit protection is automatically extracted by AI based on vehicle vibration data and the like during daily driving, the railway can be reconstructed as a competitive transportation means.

Safety2.0 is an initiative that contributes not only to safety effects but also to productivity improvements and contributes to management. I hope that UTCS will be a successful development. We believe that this will also contribute to the dissemination of Japanese Safety2.0.

#### **Author details**

*Applied Modern Control*

Safety(x,x) is illustrated by **Figure 9**.

**4. Afterword and conclusion**

*System configuration of ATP-block system.*

*History of technological progress in a train control system and Safety(x.x).*

the essential devices constituting the system mutually exchange information as IoTs, thereby realizing advanced functions. In the case of ATP-block system, the blocking device and the interlocking device, which were previously located at the station and controlling the driving direction with the adjacent station, disappear(see **Figures 7**, **8**) The history of technological progress in a train control system and its relation to the

Development of new systems involves certification work in accordance with international standards. Especially in the train control, the train is subjected to the baptism of the standard of the reliability, availability, maintainability and safety (RAMS; IEC62278). To make this baptization smart, it avoids complications and makes the system as simple as possible. Railways are one of the social systems and have a long service life. Erroneous selects can leave the roots of the trouble. Examining and simplifying the components as much as possible improves the system's visibility and facilitates certification. Furthermore, the reliability is increased by the reduction of the amount of goods, and protection becomes unnecessary. The

**12**

**Figure 9.**

**Figure 8.**

Hideo Nakamura1 \* and Yoshihisa Saito2


\*Address all correspondence to: nakamura.hideo@nihon-u.ac.jp

© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

#### **References**

[1] Concept Safety2.0. Safety2.0 Projects of Nikkei BP. Tokyo; 2016

[2] Nakamura H. [Invited Lecture] New safety culture and manufacturing on Safety 2.0. IEICEJ Tech. Report of DC-Research Group. Vol. 116, Issue 373; 2016. pp. 47-50

[3] History of Japanese Railway Signal. Tokyo: Signal Safety Association of Japan; 1980

[4] Saitou Y, Asano A, Nakamura H, Mochizuki H, Takahashi S. Restructuring of train control system by hierarchical design. IEEJ Transactions on Electronics, Information and Systems. 2016;**136**(7):923-928

[5] Saito Y, Asano A, Nakamura H, Takahashi S. A proposal for the design of integrated train control systems capable of improving reliability and safety. In: Proceedings of the Third International Conference on Railway Technology: Research, Development and Maintenance; Civil-Comp Press, Paper70; 2016. pp. 1-11

[6] Nakamura H. How to deal with revolutions in train control systems. Engineering. 2016;**2**:380-386

**15**

**Chapter 2**

**Abstract**

Flexible Control in Nanometrology

The unceasing development of new small products has increased constantly by introducing multiple facilities in line production, reduced life cycles of new innovative products, and high-precision techniques that require automation and robotization of the nanotechnology production processes. Classic size products are made in normal series and deal little change over the years, while in the field of nanotechnology, product life cycles were shortened significantly, and series production must adapt to the market challenges. Considering the fast changes and multiple innovations in production, we propose equipment that offers a high degree of flexibility and performance for quality products. To compensate efficiently, the fluctuations may appear in production series; a flexible control system is designed to adjust production for large number of items or for various models of processing. The control equipment dedicated to nanotechnologies developed by INCDMTM Bucharest offers solutions for automation processes adapted to various operations and for quick response occurring in nano-production. A modular special design offers flexibility during the process, handling and interoperable ones, along with the possibility of changes facilitated by software that controls the entire verification

*Gheorghe Popan and Ana Elisabeta Oros Daraban*

process and parameter selection for each checked item's admissibility.

optical measuring system, laser measurement system, AFM control

**1. Introduction and research problems**

**Keywords:** measurement system, calibrating nanotechnology, nanometrology,

The rapid evolution in the field of technologies related to nanomanufacturing and nano-devices based on electrical, optic, magnetic, mechanic, chemical, and biological effects would allow measurements in specific length ranges involved. Moreover, the spectacular development of nanotechnology in recent years generated the development of new devices and smaller components, trends that have created the need to measure them by developing a new nanometrology field. For standard products, measurement and control systems and equipment have been created in hundreds of years, but for nano-metric components, new appropriate measurement systems must to be created quickly. In most cases the physical principle used to measure in the usual nano-production flow from a technical point of view does not correspond with normal measurement systems. Traditional measuring means have proved some technological limits in terms of accuracy because of the physical law constraints [1, 2, 3]. Furthermore, microsensors, transducers, and ultra-accurate machines must be calibrated or verified during production and, afterward, before reception at beneficiary, because it is through them that the measuring unit is transmitted to dedicated users, meaning final producers [4].

#### **Chapter 2**

## Flexible Control in Nanometrology

*Gheorghe Popan and Ana Elisabeta Oros Daraban*

#### **Abstract**

The unceasing development of new small products has increased constantly by introducing multiple facilities in line production, reduced life cycles of new innovative products, and high-precision techniques that require automation and robotization of the nanotechnology production processes. Classic size products are made in normal series and deal little change over the years, while in the field of nanotechnology, product life cycles were shortened significantly, and series production must adapt to the market challenges. Considering the fast changes and multiple innovations in production, we propose equipment that offers a high degree of flexibility and performance for quality products. To compensate efficiently, the fluctuations may appear in production series; a flexible control system is designed to adjust production for large number of items or for various models of processing. The control equipment dedicated to nanotechnologies developed by INCDMTM Bucharest offers solutions for automation processes adapted to various operations and for quick response occurring in nano-production. A modular special design offers flexibility during the process, handling and interoperable ones, along with the possibility of changes facilitated by software that controls the entire verification process and parameter selection for each checked item's admissibility.

**Keywords:** measurement system, calibrating nanotechnology, nanometrology, optical measuring system, laser measurement system, AFM control

#### **1. Introduction and research problems**

The rapid evolution in the field of technologies related to nanomanufacturing and nano-devices based on electrical, optic, magnetic, mechanic, chemical, and biological effects would allow measurements in specific length ranges involved. Moreover, the spectacular development of nanotechnology in recent years generated the development of new devices and smaller components, trends that have created the need to measure them by developing a new nanometrology field. For standard products, measurement and control systems and equipment have been created in hundreds of years, but for nano-metric components, new appropriate measurement systems must to be created quickly. In most cases the physical principle used to measure in the usual nano-production flow from a technical point of view does not correspond with normal measurement systems. Traditional measuring means have proved some technological limits in terms of accuracy because of the physical law constraints [1, 2, 3]. Furthermore, microsensors, transducers, and ultra-accurate machines must be calibrated or verified during production and, afterward, before reception at beneficiary, because it is through them that the measuring unit is transmitted to dedicated users, meaning final producers [4].

**14**

*Applied Modern Control*

of Nikkei BP. Tokyo; 2016

[1] Concept Safety2.0. Safety2.0 Projects

[2] Nakamura H. [Invited Lecture] New safety culture and manufacturing on Safety 2.0. IEICEJ Tech. Report of DC-Research Group. Vol. 116, Issue 373;

[3] History of Japanese Railway Signal. Tokyo: Signal Safety Association of

[4] Saitou Y, Asano A, Nakamura H,

[5] Saito Y, Asano A, Nakamura H, Takahashi S. A proposal for the design of integrated train control systems capable of improving reliability and safety. In: Proceedings of the Third International Conference on Railway Technology: Research, Development and Maintenance; Civil-Comp Press,

[6] Nakamura H. How to deal with revolutions in train control systems.

Engineering. 2016;**2**:380-386

Restructuring of train control system by hierarchical design. IEEJ Transactions on Electronics, Information and Systems. 2016;**136**(7):923-928

Mochizuki H, Takahashi S.

Paper70; 2016. pp. 1-11

**References**

2016. pp. 47-50

Japan; 1980

Control and measurement techniques in nanotechnologies face specific challenges at the actual incipient stage and form tolerances of the nano-products exceeding actual measurement equipments and standards, and new generation of performant electromechanical systems is required in the field of nanometrics [5, 6]. Thus, innovative devices based on new measurement principles have been used and developed. Industrial production implies increasing manufacturing speeds on the one hand and increasing accuracy of manufacturing on the other. This can be achieved by automating and robotizing both production and production control.

Different industries developed new innovative products or materials involved that currently utilize nanotechnology. The nanoscale analysis of biosystems and of specific materials started years ago (beginning of the twentieth century) when chemistry and physics allowed small-scale characterization (bacteria, fungi studies). Recent development of medicine applications, nano-characteristics of drugs or nano-surgery, has generated advanced progress in engineering building new nanoscale systems and creating new nano-technics [7].

Other areas of emerging technologies include semiconductors and optoelectronic design and production, which increase the progress of information and communication technologies (ICT). More and more positive results engaged new initiatives and contributed to develop nanotechnology applications for structures smaller than 100 nm. Actual growth of semiconductor industry exploded toward nanotechnology boost and industrial demand raised in the last few years, generating unstable economic expansion for electronic devices in term of quality.

New nanotechnologies penetrated globally in large areas, from electronics to optical devices and from new materials to biological systems, considering upgrade of specific and customized makers offering optimal and functional parameters of the new products. This is further relevant conceiving nano-systems based on optical, electronical, mechanical, and biological nano-devices [8].

In Romania and widely, we only find significant research and innovation projects for nano-systems and nanometrology reaching the TLR 3–4 level, stage that needs upscaling to TLR 6–8. Further, industrial nano-production needs calibrated applications and metrological infrastructure at nano-dimensions to be scaled up from laboratory stage to industrial systems, which follow the quality parameters of the production flow for every relevant process [9].

Evolving toward precise production, innovations are required for efficient production structure of control systems by designing them for accreditation; thus, some procedures ask for specific parameters that are necessary to be checked.

Nanoscale dimensional accuracy covers a narrow range of tolerances. Industrial systems in nano-production can't detect smaller deviations beyond the normal tolerances, and that may have unpleasant effects by damaging the production systems. Any nano-production system requires rigorous control and verification procedure based on dimensional checking; the field of nanometrology is not developed accordingly [10].

Research and innovation in nanometrology expand the number of interested scholars, who will be supporting widely new sustainable production of nanodevices, nano-systems, and nano-materials.

Industrial processes, from medical devices to aeronautics, involve a structure where process accuracy and product quality are supervised by a system of characteristic control for every landmark product, ensuring interchangeability of product parts and the functional parameters of the product [11].

Only a few organizations have integrated this kind of research; most applications are limited at laboratory findings. The main barrier of using nanotechnology control at large industrial scale is the lack of specific infrastructure; for that reason this study proposes some solutions (**Figure 1**) [8–10].

**17**

approached.

**Figure 1.**

*Nano-electronics devices.*

**2. Experimental model**

nano-devices (**Figure 2**).

*Flexible Control in Nanometrology*

*DOI: http://dx.doi.org/10.5772/intechopen.80425*

machines, and precision tool machines) [12].

The equipment for nanometrology further presented is based on the experience of more than 30 years in research and didactic activity of the main author in the field of measuring devices and dimensional control systems. Activity in the field began with the design of control systems and devices from precision mechanics, with laser measuring and controlling equipment (laser probe heads, laser beam-scanning measuring systems, 3D cordless measuring machines, laser head, laser camshaft measuring equipment, laser calibration of coordinate measuring

The advantages of studying and realizing these systems are the basis for the next research and innovative solutions for nano-industries in terms of quality and precise manufacturing. The disadvantages are that systems and benchmarks to be verified in nanotechnological production are not palpable and in most cases are easily deformable and only distinguishable by a microscope. The transition from metrology to nanometrology required a new approach. Touch contact systems can no longer be used; the appropriate optical measurement principle—video inspection, laser scanning, and atomic force microscope (AFM) testing—must be

The equipment is designed and developed by a multidisciplinary team from INCDMTM (National Institute for Research Development for Mechatronics and Measurement Technique) in Bucharest. The equipment is mainly driven by the need to control the production flow of a recognized mobile phone company based in Romania. Meanwhile, the mobile phone company ended its tax-free period and

The chapter is structured according to primarily an introduction which highlighted the state of the art of this theme and secondly to describing the main parts of the equipment structure (experimental model). This includes Subchapter 2.1 that shows the optoelectronic control system including the charge-coupled device (CCD) camera, with examples of controllable nano-sensors; Subchapter 2.2 presents briefly the following control station with the laser control system; and then in Subchapter 2.3, the control station with the atomic force microscope (AFM) is

The experimental model presented in this paper for an innovative control and calibration equipment is built based on rotary feeding systems including table supports which are installed very precisely holding the nano-devices that need to be verified and calibrated. The equipment design allows calibration for a series of electronic nano-devices, bio-nano-devices, nano-materials, nano-sensors, or other

shown, followed by a brief conclusion and direction for future research.

relocated its production from Romania to another country.

#### *Flexible Control in Nanometrology DOI: http://dx.doi.org/10.5772/intechopen.80425*

**Figure 1.** *Nano-electronics devices.*

*Applied Modern Control*

Control and measurement techniques in nanotechnologies face specific challenges

Different industries developed new innovative products or materials involved that currently utilize nanotechnology. The nanoscale analysis of biosystems and of specific materials started years ago (beginning of the twentieth century) when chemistry and physics allowed small-scale characterization (bacteria, fungi studies). Recent development of medicine applications, nano-characteristics of drugs or nano-surgery, has generated advanced progress in engineering building new

Other areas of emerging technologies include semiconductors and optoelectronic design and production, which increase the progress of information and communication technologies (ICT). More and more positive results engaged new initiatives and contributed to develop nanotechnology applications for structures smaller than 100 nm. Actual growth of semiconductor industry exploded toward nanotechnology boost and industrial demand raised in the last few years, generat-

New nanotechnologies penetrated globally in large areas, from electronics to optical devices and from new materials to biological systems, considering upgrade of specific and customized makers offering optimal and functional parameters of the new products. This is further relevant conceiving nano-systems based on opti-

In Romania and widely, we only find significant research and innovation projects for nano-systems and nanometrology reaching the TLR 3–4 level, stage that needs upscaling to TLR 6–8. Further, industrial nano-production needs calibrated applications and metrological infrastructure at nano-dimensions to be scaled up from laboratory stage to industrial systems, which follow the quality parameters of

Evolving toward precise production, innovations are required for efficient production structure of control systems by designing them for accreditation; thus, some procedures ask for specific parameters that are necessary to be checked.

Nanoscale dimensional accuracy covers a narrow range of tolerances. Industrial systems in nano-production can't detect smaller deviations beyond the normal tolerances, and that may have unpleasant effects by damaging the production systems. Any nano-production system requires rigorous control and verification procedure based on dimensional checking; the field of nanometrology is not developed

Research and innovation in nanometrology expand the number of interested scholars, who will be supporting widely new sustainable production of nano-

Industrial processes, from medical devices to aeronautics, involve a structure where process accuracy and product quality are supervised by a system of characteristic control for every landmark product, ensuring interchangeability of product

Only a few organizations have integrated this kind of research; most applications are limited at laboratory findings. The main barrier of using nanotechnology control at large industrial scale is the lack of specific infrastructure; for that reason

ing unstable economic expansion for electronic devices in term of quality.

cal, electronical, mechanical, and biological nano-devices [8].

the production flow for every relevant process [9].

devices, nano-systems, and nano-materials.

parts and the functional parameters of the product [11].

this study proposes some solutions (**Figure 1**) [8–10].

at the actual incipient stage and form tolerances of the nano-products exceeding actual measurement equipments and standards, and new generation of performant electromechanical systems is required in the field of nanometrics [5, 6]. Thus, innovative devices based on new measurement principles have been used and developed. Industrial production implies increasing manufacturing speeds on the one hand and increasing accuracy of manufacturing on the other. This can be achieved by automat-

ing and robotizing both production and production control.

nanoscale systems and creating new nano-technics [7].

**16**

accordingly [10].

The equipment for nanometrology further presented is based on the experience of more than 30 years in research and didactic activity of the main author in the field of measuring devices and dimensional control systems. Activity in the field began with the design of control systems and devices from precision mechanics, with laser measuring and controlling equipment (laser probe heads, laser beam-scanning measuring systems, 3D cordless measuring machines, laser head, laser camshaft measuring equipment, laser calibration of coordinate measuring machines, and precision tool machines) [12].

The advantages of studying and realizing these systems are the basis for the next research and innovative solutions for nano-industries in terms of quality and precise manufacturing. The disadvantages are that systems and benchmarks to be verified in nanotechnological production are not palpable and in most cases are easily deformable and only distinguishable by a microscope. The transition from metrology to nanometrology required a new approach. Touch contact systems can no longer be used; the appropriate optical measurement principle—video inspection, laser scanning, and atomic force microscope (AFM) testing—must be approached.

The equipment is designed and developed by a multidisciplinary team from INCDMTM (National Institute for Research Development for Mechatronics and Measurement Technique) in Bucharest. The equipment is mainly driven by the need to control the production flow of a recognized mobile phone company based in Romania. Meanwhile, the mobile phone company ended its tax-free period and relocated its production from Romania to another country.

The chapter is structured according to primarily an introduction which highlighted the state of the art of this theme and secondly to describing the main parts of the equipment structure (experimental model). This includes Subchapter 2.1 that shows the optoelectronic control system including the charge-coupled device (CCD) camera, with examples of controllable nano-sensors; Subchapter 2.2 presents briefly the following control station with the laser control system; and then in Subchapter 2.3, the control station with the atomic force microscope (AFM) is shown, followed by a brief conclusion and direction for future research.

#### **2. Experimental model**

The experimental model presented in this paper for an innovative control and calibration equipment is built based on rotary feeding systems including table supports which are installed very precisely holding the nano-devices that need to be verified and calibrated. The equipment design allows calibration for a series of electronic nano-devices, bio-nano-devices, nano-materials, nano-sensors, or other nano-devices (**Figure 2**).

#### **Figure 2.** *Variety of Nano-devices necessary to be verified.*

**Figure 3.** *Optoelectronic measuring and calibration system assisted by laser and AFM control.*

The very thin nano-device calibration requires dedicated operational procedure for handling, and it is using support parts manipulated by a precision linear displacement system. These systems transfer the nano-devices by specialized automatic options (robot) to different precision measurement systems—optoelectronic, laser, or AFM—for calibration [3].

As shown in **Figure 3**, the equipment comprises a rotary feeding system, on top of which are placed eight support tables. On each support table, there is a specific plate support where nano-devices are introduced to follow the calibration procedure.

Experimental equipment includes mechanical, optical, and optoelectronic sub-ensembles, the optoelectronic measurement sub-ensembles, and the algorithms related to (real time) measurement system data acquisition, data processing, and measurement protocol presentation.

Control equipment (**Figure 4**) uses a rotary handling system including a feeding robot manipulator, a precision linear moving system, an optical measuring system, a laser measuring system, and a measuring system equipped with an atomic force microscope (AFM) [3].

The technical features of this experimental model ensure the following precision by:

**19**

boxes (**Figure 5**).

**Figure 4.**

**Figure 5.**

*Optoelectronic control.*

*The experimental equipment model.*

*Flexible Control in Nanometrology*

*DOI: http://dx.doi.org/10.5772/intechopen.80425*

Calibration procedure secures the nano-device optimal positioning in the support dedicated plate, which is precisely fixed on the support table from the rotary feeding system by computer coordination and transfer manipulator, which allows adjustments of the table during the precision displacement. For measuring operations with an AFM, a special feeding robot is used in order to keep accurate calibration characteristics. Dedicated computer programming procedure of the calibration system decides if the device is accepted (qualifying as good) continuing the production flow or is rejected (is not respecting the quality required) to scrap

The nanotechnological process can be adjusted using this equipment by programming automatic calibration for one, two, or three posts from eight available table supports [13]. The control software is versatile ordering automatic calibration process or manually controlled by a computer system or using touch screen applications.


*Applied Modern Control*

*Variety of Nano-devices necessary to be verified.*

**Figure 2.**

**Figure 3.**

laser, or AFM—for calibration [3].

measurement protocol presentation.

• Laser measurement resolution, 1 nm

• AFM characterization resolution, less than 0.5 nm

• Optoelectronic measurement resolution, 10 nm

microscope (AFM) [3].

systems, 0.2 nm

The very thin nano-device calibration requires dedicated operational procedure for handling, and it is using support parts manipulated by a precision linear displacement system. These systems transfer the nano-devices by specialized automatic options (robot) to different precision measurement systems—optoelectronic,

*Optoelectronic measuring and calibration system assisted by laser and AFM control.*

As shown in **Figure 3**, the equipment comprises a rotary feeding system, on top of which are placed eight support tables. On each support table, there is a specific plate support where nano-devices are introduced to follow the calibration procedure. Experimental equipment includes mechanical, optical, and optoelectronic sub-ensembles, the optoelectronic measurement sub-ensembles, and the algorithms related to (real time) measurement system data acquisition, data processing, and

Control equipment (**Figure 4**) uses a rotary handling system including a feeding robot manipulator, a precision linear moving system, an optical measuring system, a laser measuring system, and a measuring system equipped with an atomic force

The technical features of this experimental model ensure the following precision by:

• Displacement accuracy of the ultra-accurate-controlled linear positioning

**18**

#### **Figure 4.** *The experimental equipment model.*

**Figure 5.** *Optoelectronic control.*

Calibration procedure secures the nano-device optimal positioning in the support dedicated plate, which is precisely fixed on the support table from the rotary feeding system by computer coordination and transfer manipulator, which allows adjustments of the table during the precision displacement. For measuring operations with an AFM, a special feeding robot is used in order to keep accurate calibration characteristics. Dedicated computer programming procedure of the calibration system decides if the device is accepted (qualifying as good) continuing the production flow or is rejected (is not respecting the quality required) to scrap boxes (**Figure 5**).

The nanotechnological process can be adjusted using this equipment by programming automatic calibration for one, two, or three posts from eight available table supports [13]. The control software is versatile ordering automatic calibration process or manually controlled by a computer system or using touch screen applications.

The equipment developed in research institute INCDMTM is equipped to perform the flow control by means of three specialized systems:


In this chapter the integrated control processes for nano-production flow using these three dedicated systems are presented summarily.

#### **2.1 Optoelectronic control (microscope with CCD camera)**

This testing and controlling method permits to check up the quality and cohesion of different nano-devices: semiconductor devices (SMD discrete components), microelectronic circuits, micromachined circuits, printed microcircuits, microsensors, and transducers. The optoelectronic control method using CCD camera may be adopted for finding defects from handling, assembling, or encapsulating all types of devices listed above [3] (**Figure 6**).

The equipment used in this control process must be able to demonstrate the quality conditions of the devices mentioned in accordance with the requirements envisaged in the product design.

Equipment should include optics (optical microscope) with a magnification range of 1.5–20 X with a view area accessible and large enough for determination of details. The control procedure sets up the devices that will be examined by producers at established magnifications ranging from 1.5 to 20 X. Measurements of dimensions (length/width/diameter) will be made using the order of the same range of magnification that provide good accuracy of measurements.

The dimensional measurements with optoelectronic microscope are ranked (width, length routes, or contacts) in the range of 10–2000 μm, and measurement resolution is 10 nm. The system offers rigorous linear movement of the sample nano-device based on two perpendicular directions at a distance of at least 5 mm (matching the test plan). In this case, measuring resolution should be less than 100 nm (**Figure 7**).

Applications that require the optical control are verification of integrated circuits, verification of printed microcircuits, and verification of microsensors based on amorphous magnetic materials.

**21**

*Flexible Control in Nanometrology*

shown (**Figures 8**–**13**).

*Microprocessed circuits.*

**Figure 7.**

• Verification of junctions and contacts

a circuit within the prescribed limit

• Dimensional component control

(**Figures 12** and **13**) procedure includes:

• Control of each sensor component.

• Dimensional sensor control.

• Control of the geometric shape of the circuit

• Verification of profiles

• Control of circuit breaks

*DOI: http://dx.doi.org/10.5772/intechopen.80425*

Some examples of the optoelectronic control applications in microelectronic circuits, micromachined circuits, printed microcircuits, and microsensors are

• Verification of routes and establishing dimensional variations from the geom-

Verification of printed microcircuit (**Figures 10** and **11**) procedure includes:

• Controlling the framing of deviations from the theoretical geometric shape of

• Control of the presence and correct positioning of the components on the circuit

Verification of microsensors based on amorphous magnetic material

• Control the correct positioning for each sensor on the circuit.

• Control the alignment of each sensor in the circuit.

Verification of integrated circuit (**Figures 8** and **9**) procedure includes:

etry of the proposed design by comparing with a theoretical form

**Figure 6.** *Integrated circuit.*

**Figure 7.** *Microprocessed circuits.*

*Applied Modern Control*

• Laser control

• AFM control

The equipment developed in research institute INCDMTM is equipped to

In this chapter the integrated control processes for nano-production flow using

This testing and controlling method permits to check up the quality and cohesion of different nano-devices: semiconductor devices (SMD discrete components), microelectronic circuits, micromachined circuits, printed microcircuits, microsensors, and transducers. The optoelectronic control method using CCD camera may be adopted for finding defects from handling, assembling, or encapsulating all

The equipment used in this control process must be able to demonstrate the quality conditions of the devices mentioned in accordance with the requirements

Equipment should include optics (optical microscope) with a magnification range of 1.5–20 X with a view area accessible and large enough for determination of details. The control procedure sets up the devices that will be examined by producers at established magnifications ranging from 1.5 to 20 X. Measurements of dimensions (length/width/diameter) will be made using the order of the same

The dimensional measurements with optoelectronic microscope are ranked (width, length routes, or contacts) in the range of 10–2000 μm, and measurement resolution is 10 nm. The system offers rigorous linear movement of the sample nano-device based on two perpendicular directions at a distance of at least 5 mm (matching the test plan). In

Applications that require the optical control are verification of integrated circuits, verification of printed microcircuits, and verification of microsensors based

range of magnification that provide good accuracy of measurements.

this case, measuring resolution should be less than 100 nm (**Figure 7**).

perform the flow control by means of three specialized systems:

• Optoelectronic control (microscope with CCD camera)

these three dedicated systems are presented summarily.

types of devices listed above [3] (**Figure 6**).

envisaged in the product design.

on amorphous magnetic materials.

**2.1 Optoelectronic control (microscope with CCD camera)**

**20**

**Figure 6.** *Integrated circuit.*

Some examples of the optoelectronic control applications in microelectronic circuits, micromachined circuits, printed microcircuits, and microsensors are shown (**Figures 8**–**13**).

Verification of integrated circuit (**Figures 8** and **9**) procedure includes:


Verification of printed microcircuit (**Figures 10** and **11**) procedure includes:


Verification of microsensors based on amorphous magnetic material (**Figures 12** and **13**) procedure includes:


**Figure 8.** *Microelectronic circuits.*

**Figure 9.** *Micro-imprinted circuits.*

**Figure 10.** *Microcircuits.*

#### **2.2 Laser control**

Laser measurement technologies gradually developed using multiple measurement principles that allow a large control flexibility and applicability for measurement and checking procedures.

**23**

**Figure 13.** *Inductive sensor.*

**Figure 12.**

*Cable routes and metal-plated holes.*

**Figure 11.** *Microcircuit.*

keeping formation flying.

Laser telemetry measurement principle offers a great variability of distance measuring systems up to kilometer lengths. The INCDMTM center developed applications to measure distance by telemetry satellites during the formation flying useful to maintain and adjust flying positioning. This application allows monitoring of distance length between satellites, and it controls trajectory of each satellite for

*Flexible Control in Nanometrology*

*DOI: http://dx.doi.org/10.5772/intechopen.80425*

*Flexible Control in Nanometrology DOI: http://dx.doi.org/10.5772/intechopen.80425*

**Figure 11.** *Microcircuit.*

*Applied Modern Control*

**22**

**2.2 Laser control**

**Figure 10.** *Microcircuits.*

**Figure 9.**

**Figure 8.**

*Microelectronic circuits.*

*Micro-imprinted circuits.*

ment and checking procedures.

Laser measurement technologies gradually developed using multiple measurement principles that allow a large control flexibility and applicability for measure-

**Figure 12.** *Cable routes and metal-plated holes.*

**Figure 13.** *Inductive sensor.*

Laser telemetry measurement principle offers a great variability of distance measuring systems up to kilometer lengths. The INCDMTM center developed applications to measure distance by telemetry satellites during the formation flying useful to maintain and adjust flying positioning. This application allows monitoring of distance length between satellites, and it controls trajectory of each satellite for keeping formation flying.

#### *Applied Modern Control*

Interferometry principle is used for measurement covering distances of 80–120 meters and allows high resolution up to 0.01 nm. With a large experience in the development of checking and measurement applications for sensors, transducers, coordinate measuring machines, and precise CNC machines, we proposed to use the interferometry principle for nanotechnology processes where very precise displacements in a network system can be supervised with specific sensors. Laser triangulation is an accurate measurement principle with resolution of 1 nm. This method uses measurement referential to a point for distance and object presence determinations or referential to a line covering 3D forms and a profile dimension.

The new equipment designed for very precise measurement within time checking methods applies triangulation principle. Nevertheless our institute developed measuring systems using triangulation method 30 years ago [1]; in this case to assemble this measuring equipment on the nano-production flow, we acquired some measuring systems from a specialized company.

The purpose of using this method of control is to check the quality conditions of the semiconductor devices (discrete component-type SMD) of microsensors and transducers (e.g., control surfaces, movement control, control distance/size, position control, etc.).

The equipment used in this control must be able to demonstrate the quality conditions of the devices mentioned in accordance with the requirements envisaged in the design. It should include laser equipment and devices enabling precision movements on three axes (nano-positioning stage).

The measuring principle is the method of triangulation, having a measuring range of 5 mm, measurement resolution of 1 nm, and laser measurement resolution of 1 nm.

The value of the dimensions (width/length/height of routes, etc.) that can be checked is in the range 1–2000 μm. The system allows linear movement of the sample in three directions perpendicular to distances of at least 5 mm with nanometer precision. The precise positioning table is fixed in the laser calibration position as shown in **Figure 14**.

One of the applications appropriate for using laser-based measuring method is verification of integrated circuits, procedure that includes:


Some examples of electronic microcircuits where laser control is applicable are shown in **Figures 15**–**17**. First, one control application defined by triangulation method for measuring and verification of the profile, positioning, and present splice of a pin in the integrated circuit is presented in **Figure 15**.

In microcircuit manufacturing, one important issue raised by specialists is the presence and correct checking position of each specific component to ensure the designed function of integrated circuits. This checking is presented in **Figure 16**(**a**) using a Keyence scanner. Continuous trends of minimizing the characteristic dimensions in integrated circuits and the rapid multiplication of functions determined for the same products lead to specialized very fine and narrow circuits' paths. Each circuit's path must to be produced respecting some

**25**

**Figure 16.**

**Figure 14.**

**Figure 15.**

*Verification of the profile of pins for an integrated circuit.*

*Verification of the profile of microcircuits (a) and integrated circuits (b).*

*Calibration with laser.*

*Flexible Control in Nanometrology*

*DOI: http://dx.doi.org/10.5772/intechopen.80425*

rigorous requirements: minimum dimensions, distance between two of the closed paths, and the transversal profile of each path line. If the checking of circuit's form and the correspondence with theoretical design of the final product is realized with optoelectronic methods using video inspection, the verification of integrated circuit

parameters is made by laser triangulation as shown in **Figure 16**(**b**).

#### *Flexible Control in Nanometrology DOI: http://dx.doi.org/10.5772/intechopen.80425*

*Applied Modern Control*

profile dimension.

position control, etc.).

as shown in **Figure 14**.

of 1 nm.

Interferometry principle is used for measurement covering distances of 80–120 meters and allows high resolution up to 0.01 nm. With a large experience in the development of checking and measurement applications for sensors, transducers, coordinate measuring machines, and precise CNC machines, we proposed to use the interferometry principle for nanotechnology processes where very precise displacements in a network system can be supervised with specific sensors. Laser triangulation is an accurate measurement principle with resolution of 1 nm. This method uses measurement referential to a point for distance and object presence determinations or referential to a line covering 3D forms and a

The new equipment designed for very precise measurement within time checking methods applies triangulation principle. Nevertheless our institute developed measuring systems using triangulation method 30 years ago [1]; in this case to assemble this measuring equipment on the nano-production flow, we acquired

The purpose of using this method of control is to check the quality conditions of the semiconductor devices (discrete component-type SMD) of microsensors and transducers (e.g., control surfaces, movement control, control distance/size,

The equipment used in this control must be able to demonstrate the quality conditions of the devices mentioned in accordance with the requirements envisaged in the design. It should include laser equipment and devices enabling precision

The measuring principle is the method of triangulation, having a measuring range of 5 mm, measurement resolution of 1 nm, and laser measurement resolution

The value of the dimensions (width/length/height of routes, etc.) that can be checked is in the range 1–2000 μm. The system allows linear movement of the sample in three directions perpendicular to distances of at least 5 mm with nanometer precision. The precise positioning table is fixed in the laser calibration position

One of the applications appropriate for using laser-based measuring method is

• Verification of routes and dimensional variations of the geometry identified to

Some examples of electronic microcircuits where laser control is applicable are shown in **Figures 15**–**17**. First, one control application defined by triangulation method for measuring and verification of the profile, positioning, and present

In microcircuit manufacturing, one important issue raised by specialists is the presence and correct checking position of each specific component to ensure the

some measuring systems from a specialized company.

movements on three axes (nano-positioning stage).

verification of integrated circuits, procedure that includes:

• Verification of junctions and contacts

• Verification of profiles

the proposed design by comparing with a theoretical form

splice of a pin in the integrated circuit is presented in **Figure 15**.

designed function of integrated circuits. This checking is presented in **Figure 16**(**a**) using a Keyence scanner. Continuous trends of minimizing the characteristic dimensions in integrated circuits and the rapid multiplication of functions determined for the same products lead to specialized very fine and narrow circuits' paths. Each circuit's path must to be produced respecting some

**24**

rigorous requirements: minimum dimensions, distance between two of the closed paths, and the transversal profile of each path line. If the checking of circuit's form and the correspondence with theoretical design of the final product is realized with optoelectronic methods using video inspection, the verification of integrated circuit parameters is made by laser triangulation as shown in **Figure 16**(**b**).

**Figure 14.** *Calibration with laser.*

**Figure 15.** *Verification of the profile of pins for an integrated circuit.*

**Figure 16.** *Verification of the profile of microcircuits (a) and integrated circuits (b).*

**Figure 17.** *Gripper clamps up the nano-device support.*

#### **2.3 AFM control**

One of the problematic issues in the nano-production line control is the automatic maneuver of nano-devices. To protect nano-devices (microcircuits) during checking operations, there are specific item supports used with automatic precise displacement. Therefore, considering the nano-device control procedures, all parameters settled for measurement can be easily provided in each control point of the new equipment (optical, laser, and especially for atomic force microscope, AFM, characterization).

The purpose of using this AFM control method for nano-device testing and inspection plays an important role, and it is appropriate to check and to keep right conditions of integrity and quality of porous alumina membranes (alumina template) having pore sizes included in the nanometer range.

This control method can be used for inspecting defects that may result from the production (manufacturing industry), handling, or assembly of alumina membranes [9]. Control equipment used must be able to demonstrate the quality conditions of the porous alumina membranes in accordance with the requirements envisaged by product theoretical design. Control equipment is endowed with specialized systems that include an AFM.

The method of verification, control, and calibration includes the following procedure characteristics:


**27**

**Figure 18.**

*Flexible Control in Nanometrology*

• Verification of profiles.

control software program.

from AFM measuring head.

position (**Figures 18** and **19**).

*DOI: http://dx.doi.org/10.5772/intechopen.80425*

• Verification of membrane integrity.

Applicable for AFM control are porous alumina membranes used for obtaining

nanowires via the electrode position process, and the procedure includes:

proposed design, compared with theoretical form.

interoperation displacements of support plates (**Figure 17**).

displacement in cantilever in relation to the beam spot.

*Robot lifting nano-device related with AFM positioning socket.*

• Verification of the dimensional pore width and of the geometry of the

Checking and control method using AFM is described summarily as follows: nano-devices are positioned on the support plate which is useful for automatic maneuver. Maneuver operations are driven by a robot that has a clapping system for

Before automatic operation offsets, it is required to set up the measuring head of AFM, which is equipped with a special support fixed on the adjustment unit of this head. The special support must be provided with guidance systems for support plate corresponding to each nano-device that must be calibrated. Adaptation of AFM measuring head position is settled following two rectangular directions using two fine-pitch screws that allow micron precise positioning of the measuring head referential to laser beam. Setting up of detection systems is performed by using

Automatic feeding of support plates with nano-devices is completed by a precise robot using a special gripper with fine claw clamps that hold and fix the support plates with nano-device in specific hole. Gripper form allows maneuver and fixing of nano-device support compatible with the guiding system of the special support

One important issue regarding AFM operation is the laser beam alignment into the cantilever. For nano-device precise positioning for calibration procedure, a universal measuring head was selected. The laser beam alignment is realized by joist

The gripper clamps up the support table with the calibration nano-device, and it is built to introduce the nano-device fixed on its support plate ready to be verified directly in the AFM socket without protection cap removal (**Figure 18**). This automatic process using AFM control admits time-saving and more productivity. The robot lifts the support with the nano-device that must be calibrated to the height of the positioning socket of the AFM and introduces that support in the right

• The measuring resolution must be higher than 0.5 nm.

#### *Flexible Control in Nanometrology DOI: http://dx.doi.org/10.5772/intechopen.80425*

*Applied Modern Control*

**2.3 AFM control**

*Gripper clamps up the nano-device support.*

**Figure 17.**

AFM, characterization).

procedure characteristics:

at least 5 mm.

One of the problematic issues in the nano-production line control is the automatic maneuver of nano-devices. To protect nano-devices (microcircuits) during checking operations, there are specific item supports used with automatic precise displacement. Therefore, considering the nano-device control procedures, all parameters settled for measurement can be easily provided in each control point of the new equipment (optical, laser, and especially for atomic force microscope,

The purpose of using this AFM control method for nano-device testing and inspection plays an important role, and it is appropriate to check and to keep right conditions of integrity and quality of porous alumina membranes (alumina tem-

This control method can be used for inspecting defects that may result from the production (manufacturing industry), handling, or assembly of alumina membranes [9]. Control equipment used must be able to demonstrate the quality conditions of the porous alumina membranes in accordance with the requirements envisaged by product theoretical design. Control equipment is endowed with

The method of verification, control, and calibration includes the following

• The scanning range on XY (sample plan) must be at least 100 × 100 μm.

• The values of dimensions (width/length, diameter pores, etc.) that can be

• The system allows motorized sample stage in three directions, at a distance of

• The device must provide noncontact imaging solutions for nanoscale metrology.

plate) having pore sizes included in the nanometer range.

• The scanning range on Z must be at least 25 μm.

• The measuring resolution must be higher than 0.5 nm.

checked are in the range of 1–500 nm.

specialized systems that include an AFM.

**26**

Applicable for AFM control are porous alumina membranes used for obtaining nanowires via the electrode position process, and the procedure includes:


Checking and control method using AFM is described summarily as follows: nano-devices are positioned on the support plate which is useful for automatic maneuver. Maneuver operations are driven by a robot that has a clapping system for interoperation displacements of support plates (**Figure 17**).

Before automatic operation offsets, it is required to set up the measuring head of AFM, which is equipped with a special support fixed on the adjustment unit of this head. The special support must be provided with guidance systems for support plate corresponding to each nano-device that must be calibrated. Adaptation of AFM measuring head position is settled following two rectangular directions using two fine-pitch screws that allow micron precise positioning of the measuring head referential to laser beam. Setting up of detection systems is performed by using control software program.

Automatic feeding of support plates with nano-devices is completed by a precise robot using a special gripper with fine claw clamps that hold and fix the support plates with nano-device in specific hole. Gripper form allows maneuver and fixing of nano-device support compatible with the guiding system of the special support from AFM measuring head.

One important issue regarding AFM operation is the laser beam alignment into the cantilever. For nano-device precise positioning for calibration procedure, a universal measuring head was selected. The laser beam alignment is realized by joist displacement in cantilever in relation to the beam spot.

The gripper clamps up the support table with the calibration nano-device, and it is built to introduce the nano-device fixed on its support plate ready to be verified directly in the AFM socket without protection cap removal (**Figure 18**). This automatic process using AFM control admits time-saving and more productivity.

The robot lifts the support with the nano-device that must be calibrated to the height of the positioning socket of the AFM and introduces that support in the right position (**Figures 18** and **19**).

**Figure 18.** *Robot lifting nano-device related with AFM positioning socket.*

#### **Figure 19.**

*Robot positioning nano-device support plate in the AFM socket.*

**Figure 20.** *Robotic performance of AFM calibration.*

**29**

**Figure 23.**

*Nanowires in the porous alumina membrane (AFM image).*

*Flexible Control in Nanometrology*

calibration position (**Figure 20**).

production flow.

**Figure 22.**

*DOI: http://dx.doi.org/10.5772/intechopen.80425*

The robot clamps off the gripper and takes down the support with the device that needs to be calibrated, and the AFM catches and holds the nano-device in the

Control and checking technology defines if the verification is done in each checking position or only in a few checking positions and ensures the monitoring of settled characteristics important to be verified in every checking position. After every finalized checking operation, the equipment decides if the nanodevice is within the tolerance limits and if it is accepted or not. If a specific characteristic of every nano-device is not corresponding within the settled limits of the technology that is designated by the measurement program, the feeding robot receives the command REBUT in every checking position, and the nano-device is eliminated from the production flow [13]. If the nano-device is accepted according to settled limits for each checking position, the robot receives the command GOOD, and the nano-device is introduced ensuing further into

*Nanowires obtained through the process of nano-disposition in porous alumina membranes (SEM images).*

#### *Flexible Control in Nanometrology DOI: http://dx.doi.org/10.5772/intechopen.80425*

*Applied Modern Control*

**28**

**Figure 21.**

*Porous alumina membranes with pore diameters at nanoscale (SEM image) (a) top view and (b) lateral view.*

**Figure 20.**

**Figure 19.**

*Robotic performance of AFM calibration.*

*Robot positioning nano-device support plate in the AFM socket.*

The robot clamps off the gripper and takes down the support with the device that needs to be calibrated, and the AFM catches and holds the nano-device in the calibration position (**Figure 20**).

Control and checking technology defines if the verification is done in each checking position or only in a few checking positions and ensures the monitoring of settled characteristics important to be verified in every checking position. After every finalized checking operation, the equipment decides if the nanodevice is within the tolerance limits and if it is accepted or not. If a specific characteristic of every nano-device is not corresponding within the settled limits of the technology that is designated by the measurement program, the feeding robot receives the command REBUT in every checking position, and the nano-device is eliminated from the production flow [13]. If the nano-device is accepted according to settled limits for each checking position, the robot receives the command GOOD, and the nano-device is introduced ensuing further into production flow.

**Figure 22.**

*Nanowires obtained through the process of nano-disposition in porous alumina membranes (SEM images).*

**Figure 23.** *Nanowires in the porous alumina membrane (AFM image).*

Some relevant pictures present examples of AFM control applicability to porous alumina membranes where pore diameters are in the range of nanometers (**Figures 21**–**23**).

The images from **Figures 21**–**23** are achieved in the National Institute of Research and Development for Technical Physics, Iasi, Romania, during the collaborative research using scanning electron microscope (SEM) and AFM. These applications are demonstrative for AFM characterization and are dedicated for very precise control and checking processes in the nano-production flow.

#### **3. Conclusions and future research**

The experimental model permits optical, laser, and AFM microscopic verifications of realized nano-devices in order to correct possible production errors [3, 6, 12, 13]; thus, it allows nano-production calibration and automatic selection of rebuttal during flow processes for dedicated dimensional control in range from less than 1 nm up to micrometers or millimeters.

Future research will aim at the development of detailed technologies for various applications in nano-device production field that need to be calibrated covering all ranges of electronic nano-devices, optical nano-devices, biological nano-devices, nano-materials, and nano-sensors.

To evolve from laboratory stage to nanotechnology production lines, more research and innovations may allow over passing the actual barriers:


Nanometrology opens opportunity creation of international standards and equipments for calibration of the products and equipments used in industrial production and offers more chances of new scientific discoveries regarding innovative commercial products.

The future development of nanotechnology cannot be achieved without progress in ensuring a well-controlled, stable production carrying dimensional control and in terms of other quality characteristics. This depends both on the strategy of each area of development and especially on the joint development of the nanotechnology field. First, research should be coordinated and developed in collaboration with companies, and secondly research for standardization in the field of nanometrology must be promoted by government programs. Efforts need to be united between those with common concerns for the progress of the nanotechnologies in precise industry.

For unitary development and interchangeable products, rules and standards need to be created at European and international status both for the acceptance of

**31**

**Author details**

provided the original work is properly cited.

Gheorghe Popan\* and Ana Elisabeta Oros Daraban

\*Address all correspondence to: popangeorge@yahoo.com

Technique (INCDMTM), Bucharest, Romania

*Flexible Control in Nanometrology*

*DOI: http://dx.doi.org/10.5772/intechopen.80425*

and vibration not only of temperature and humidity.

© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium,

National Institute of Research and Development in Mechatronics and Measurement

production and for systematization of products and components at the nanoscale. For large-scale production of nanotechnology products, equipments must be developed for both industrial production and production control. In order to ensure stable production, international rules must be developed and immerged for calibration of production flow and for calibration of control equipment production. Another issue that needs to be considered is that of environmental production conditions. We need to rethink environmental standards for this type of production. The old classification and standardization of clean rooms no longer correspond, and it is necessary to improve the clean room technical standards and add specific parameters. At the atomic force microscope (AFM), the measured parameter value is drastically influenced by its position relative to the air circulation system, noise,

#### *Flexible Control in Nanometrology DOI: http://dx.doi.org/10.5772/intechopen.80425*

*Applied Modern Control*

(**Figures 21**–**23**).

Some relevant pictures present examples of AFM control applicability to porous alumina membranes where pore diameters are in the range of nanometers

The images from **Figures 21**–**23** are achieved in the National Institute of Research and Development for Technical Physics, Iasi, Romania, during the collaborative research using scanning electron microscope (SEM) and AFM. These applications are demonstrative for AFM characterization and are dedicated for very

The experimental model permits optical, laser, and AFM microscopic verifica-

[3, 6, 12, 13]; thus, it allows nano-production calibration and automatic selection of rebuttal during flow processes for dedicated dimensional control in range from less

Future research will aim at the development of detailed technologies for various applications in nano-device production field that need to be calibrated covering all ranges of electronic nano-devices, optical nano-devices, biological nano-devices,

To evolve from laboratory stage to nanotechnology production lines, more

• Traditional measurement techniques used for normal dimensional character-

• Special rules and standards must be introduced for nano-structures and nanomaterial characterization reducing errors in inspection and quality checking

• More innovative equipments must be projected in order to solve the mentioned

• Different specific studies for new equipments for control production regarding nano-structure proprieties should promote reproducible production of nano-

The future development of nanotechnology cannot be achieved without progress in ensuring a well-controlled, stable production carrying dimensional control and in terms of other quality characteristics. This depends both on the strategy of each area of development and especially on the joint development of the nanotechnology field. First, research should be coordinated and developed in collaboration with companies, and secondly research for standardization in the field of nanometrology must be promoted by government programs. Efforts need to be united between those with common concerns for the progress of the nanotechnologies in precise industry. For unitary development and interchangeable products, rules and standards need to be created at European and international status both for the acceptance of

Nanometrology opens opportunity creation of international standards and equipments for calibration of the products and equipments used in industrial production and offers more chances of new scientific discoveries regarding innova-

research and innovations may allow over passing the actual barriers:

ization cannot be applied to nano-structures.

tions of realized nano-devices in order to correct possible production errors

precise control and checking processes in the nano-production flow.

**3. Conclusions and future research**

than 1 nm up to micrometers or millimeters.

nano-materials, and nano-sensors.

structures and nano-materials.

procedures.

tive commercial products.

issues.

**30**

production and for systematization of products and components at the nanoscale. For large-scale production of nanotechnology products, equipments must be developed for both industrial production and production control. In order to ensure stable production, international rules must be developed and immerged for calibration of production flow and for calibration of control equipment production. Another issue that needs to be considered is that of environmental production conditions. We need to rethink environmental standards for this type of production. The old classification and standardization of clean rooms no longer correspond, and it is necessary to improve the clean room technical standards and add specific parameters. At the atomic force microscope (AFM), the measured parameter value is drastically influenced by its position relative to the air circulation system, noise, and vibration not only of temperature and humidity.

### **Author details**

Gheorghe Popan\* and Ana Elisabeta Oros Daraban National Institute of Research and Development in Mechatronics and Measurement Technique (INCDMTM), Bucharest, Romania

\*Address all correspondence to: popangeorge@yahoo.com

© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

#### **References**

[1] Popan Gh, Sorea S, Savu T. Mechatronic System for Scavenge the Laser Ray. Romanian Review Precision Mechanics, Optics & Mechatronics. 2007;**31**. ISSN 1584-5928

[2] Chiriac H, Popan G, Ursu D, Gheorghe M, The Importance of the Production Calibration in the Nanodomain. Romanian Review Precision, Mechanics Optics Mechatronics. 2008;**33**. ISSN 1584-5928

[3] Popan G, Angelescu D. Flexible control system used in the nanotechnological production flow. In: Proceedings of 5th International Conference on Advanced Manufacturing Engineering and Technologies (NEWTECH 2017), Lecture Notes in Mechanical Engineering. pp. 67-77. DOI: 10.1007/978-3-319-56430-2\_6

[4] Leão C-P, Soares F-O, Machado J-M, Seabra E, Rodrigues H. Design and development of an industrial network laboratory. International Journal of Emerging Technologies in Learning. 2011;**6**(2):21-26. DOI: 10.3991/ijet. v6iS1.1615

[5] Chiriac H, Popan G, Ursu D, Gheorghe M. Nano-technological measuring systems. In: Proceedings of the Second International Conference on Innovations, Recent Trend and Challenges in Mechatronics, Mechanical Engineering, New High-Tech Products Development (MECAHITECH `10), Bucharest. 2010. pp. 486- 491, https://incdmtm.ro/editura/ imagini/Proceedings%20of%20 MECAHITECH%202010.pdf

[6] Donmez A. EL Program: Smart Manufacturing Processes and Equipment, National Institute of Standards and Technology, US Department of Commerce. 2013; https://www.nist.gov/sites/default/ files/documents/el/isd/sbm/SMPE-ProgramDescription\_2013.pdf

[7] Machado J, Denis B, Lesage J-J. A generic approach to build plant models for DES verification purposes. In: Proceedings of the Eighth International Workshop on Discrete Event Systems, (WODES` 06); 10-12 July 2006; Michigan. pp. 407-412

[8] Sorin S, Popan G, Atanasescu A. Nanotechnology and solar cells. In: Proceedings of the Second International Conference on Innovations, Recent Trend and Challenges in Mechatronics, Mechanical Engineering, New High-Tech Products Development (MECAHITECH `10), Bucharest. 2010. pp. 473-476

[9] Barros C, Leão C-P, Soare F, Minas G, Machado J-M. RePhyS: A multidisciplinary experience in remote physiological systems laboratory. International Journal of Online Engineering. 2013;**9**(Sppl. 5):21-24. http://online-journals.org/index. php/i-joe/article/view/2756/2650

[10] Popan G, Gheorghe G, Vieru A. Control procedures in the production flow. In: Proceeding of International Conference on Innovations, Recent Trend and Challenges in Mechatronics, Mechanical Engineering, New High-Tech Products Development (MECAHITECH `09) Bucharest; 8-9 October 2009

[11] Silva M, Pereira F, Soares F, Leão C-P, Machado J-M, Carvalho V. An overview of industrial communication networks. In: Springer Cham editor, New Trends in Mechanism and Machine Science. Vol. 24; 2015. pp. 933-940. DOI: 10.1007/978-3-319-09411-3\_97

[12] Popan G, Palade D-D, Lung I, Tacutu I. Influence of constructivefunctional elements of the laser head

**33**

*Flexible Control in Nanometrology*

Management & Business

*DOI: http://dx.doi.org/10.5772/intechopen.80425*

systems accuracy. In: Proceedings of the 6th International Working Conference – "Total quality management- advanced and intelligent approaches", 7-11 June, Belgrade, Serbia Rev: 8843 Total Quality

[13] Vieira G-G, Varela ML-R, Putnik G-D, Machado J-M, Trojanowska J. Integrated platform for real-time control and production and productivity monitoring and analysis. Romanian Review Precision Mechanics, Optics and Mechatronics Review. 2016;**50**:119-127. https://search.proquest.com/openview/ 390da36d8112b1fe4783f794cf3006aa/1 ?pq-origsite=gscholar&cbl=2035039

*Flexible Control in Nanometrology DOI: http://dx.doi.org/10.5772/intechopen.80425*

systems accuracy. In: Proceedings of the 6th International Working Conference – "Total quality management- advanced and intelligent approaches", 7-11 June, Belgrade, Serbia Rev: 8843 Total Quality Management & Business

[13] Vieira G-G, Varela ML-R, Putnik G-D, Machado J-M, Trojanowska J. Integrated platform for real-time control and production and productivity monitoring and analysis. Romanian Review Precision Mechanics, Optics and Mechatronics Review. 2016;**50**:119-127. https://search.proquest.com/openview/ 390da36d8112b1fe4783f794cf3006aa/1 ?pq-origsite=gscholar&cbl=2035039

**32**

v6iS1.1615

*Applied Modern Control*

**References**

[1] Popan Gh, Sorea S, Savu T.

[2] Chiriac H, Popan G, Ursu D, Gheorghe M, The Importance of the Production Calibration in the Nanodomain. Romanian Review Precision, Mechanics Optics Mechatronics.

[3] Popan G, Angelescu D. Flexible control system used in the nanotechnological production flow. In: Proceedings of 5th International Conference on Advanced Manufacturing Engineering and Technologies (NEWTECH 2017), Lecture Notes in Mechanical Engineering. pp. 67-77. DOI: 10.1007/978-3-319-56430-2\_6

[4] Leão C-P, Soares F-O, Machado J-M, Seabra E, Rodrigues H. Design and development of an industrial network laboratory. International Journal of Emerging Technologies in Learning. 2011;**6**(2):21-26. DOI: 10.3991/ijet.

[5] Chiriac H, Popan G, Ursu D, Gheorghe M. Nano-technological measuring systems. In: Proceedings of the Second International Conference on Innovations, Recent Trend and Challenges in Mechatronics, Mechanical Engineering, New High-Tech Products Development (MECAHITECH `10), Bucharest. 2010. pp. 486- 491, https://incdmtm.ro/editura/ imagini/Proceedings%20of%20 MECAHITECH%202010.pdf

[6] Donmez A. EL Program: Smart Manufacturing Processes and Equipment, National Institute of Standards and Technology, US Department of Commerce. 2013; https://www.nist.gov/sites/default/

2007;**31**. ISSN 1584-5928

2008;**33**. ISSN 1584-5928

Mechatronic System for Scavenge the Laser Ray. Romanian Review Precision Mechanics, Optics & Mechatronics.

files/documents/el/isd/sbm/SMPE-ProgramDescription\_2013.pdf

[7] Machado J, Denis B, Lesage J-J. A generic approach to build plant models for DES verification purposes. In: Proceedings of the Eighth International Workshop on Discrete Event Systems, (WODES` 06); 10-12 July 2006;

[8] Sorin S, Popan G, Atanasescu A. Nanotechnology and solar cells. In: Proceedings of the Second International Conference on Innovations, Recent Trend and Challenges in Mechatronics,

Mechanical Engineering, New High-Tech Products Development (MECAHITECH `10), Bucharest. 2010.

[9] Barros C, Leão C-P, Soare F, Minas G, Machado J-M. RePhyS: A multidisciplinary experience in remote physiological systems laboratory. International Journal of Online Engineering. 2013;**9**(Sppl. 5):21-24. http://online-journals.org/index. php/i-joe/article/view/2756/2650

[10] Popan G, Gheorghe G, Vieru A. Control procedures in the production flow. In: Proceeding of International Conference on Innovations, Recent Trend and Challenges in Mechatronics,

[11] Silva M, Pereira F, Soares F, Leão C-P, Machado J-M, Carvalho V. An overview of industrial communication networks. In: Springer Cham editor, New Trends in Mechanism and Machine Science. Vol. 24; 2015. pp. 933-940. DOI:

10.1007/978-3-319-09411-3\_97

[12] Popan G, Palade D-D, Lung I, Tacutu I. Influence of constructivefunctional elements of the laser head

Mechanical Engineering, New High-Tech Products Development (MECAHITECH `09) Bucharest; 8-9

October 2009

Michigan. pp. 407-412

pp. 473-476

**35**

**Chapter 3**

*Ming-Sen Hu*

**1. Introduction**

**Abstract**

The Design and Development of

Control System for High Vacuum

Glove Box with Cycling Cleaning

and Regeneration

Deoxygenated and Water-Removal

This study proposed a high vacuum deoxygenated and water removal glove box control system. Through parameter setting, the system can automatically perform various glove box cleaning operations and quickly reach the micro-oxygen and microwater concentration requirements. In addition, two sets of reaction tanks are built in the system, and the hardware pipeline switching design and monitoring software control are used to provide two sets of reaction tanks to execute the cycling cleaning and cycling regeneration operation procedures synchronously, which can effectively solve the problem of interruption of the experimental process, improve the efficiency of its cleaning operations, and greatly reduce the manpower and material costs of the glove box operation. In addition, the system can automatically record the relevant data during various operations for the analysis of glove box monitoring effectiveness.

**Keywords:** glove box, control system, cycling cleaning, cycling regeneration,

A glove box, also known as an anaerobic station, is a vacuum environment without any water (H2O), oxygen (O2), and organic gas, where high purity inert gas is filled, and the active substances are filtered out, which can prevent the external personnel from direct contact, so that the materials placed in the glove box can be stored or tested in the vacuum environment free of water, gas, and oxygen, and the material is operated by people in a safe and contactless manner [1]. The glove box is used in a wide range of applications such as the scientific research of chemical/chemical engineering/material/drug, and so on; the research and manufacturing of organic optoelectronic OLED/polymer PLED light emitting displays; the manufacturing and research of lithium ion batteries/lithium polymer batteries/solar cells/high capacity capacitors; the research and manufacturing of special (HID) light bulbs; metal welding or laser metal welding process; the preservation or research and manufacturing

monitoring software, system integration, performance analysis

of noble sensitive drugs and nuclear energy research; and so on [2–4].

#### **Chapter 3**

## The Design and Development of Control System for High Vacuum Deoxygenated and Water-Removal Glove Box with Cycling Cleaning and Regeneration

*Ming-Sen Hu*

### **Abstract**

This study proposed a high vacuum deoxygenated and water removal glove box control system. Through parameter setting, the system can automatically perform various glove box cleaning operations and quickly reach the micro-oxygen and microwater concentration requirements. In addition, two sets of reaction tanks are built in the system, and the hardware pipeline switching design and monitoring software control are used to provide two sets of reaction tanks to execute the cycling cleaning and cycling regeneration operation procedures synchronously, which can effectively solve the problem of interruption of the experimental process, improve the efficiency of its cleaning operations, and greatly reduce the manpower and material costs of the glove box operation. In addition, the system can automatically record the relevant data during various operations for the analysis of glove box monitoring effectiveness.

**Keywords:** glove box, control system, cycling cleaning, cycling regeneration, monitoring software, system integration, performance analysis

#### **1. Introduction**

A glove box, also known as an anaerobic station, is a vacuum environment without any water (H2O), oxygen (O2), and organic gas, where high purity inert gas is filled, and the active substances are filtered out, which can prevent the external personnel from direct contact, so that the materials placed in the glove box can be stored or tested in the vacuum environment free of water, gas, and oxygen, and the material is operated by people in a safe and contactless manner [1]. The glove box is used in a wide range of applications such as the scientific research of chemical/chemical engineering/material/drug, and so on; the research and manufacturing of organic optoelectronic OLED/polymer PLED light emitting displays; the manufacturing and research of lithium ion batteries/lithium polymer batteries/solar cells/high capacity capacitors; the research and manufacturing of special (HID) light bulbs; metal welding or laser metal welding process; the preservation or research and manufacturing of noble sensitive drugs and nuclear energy research; and so on [2–4].

The glove box is mainly composed of a glove box body, a vacuum system, a gas circulation exchange system, and a control system [5]. The glove box body is provided with an antechamber (also called a transfer box) and an isolation glove. Usually a viewing window is arranged on the front of the body, so that the operator can clearly observe the operating conditions inside the box, and the operation process can be intuitively displayed in front of the operator. The glove box facilitates operators with isolation gloves in an anhydrous, oxygen free, and vacuum environment. Basically, the vacuum glove box is an important application of vacuum technology [6–9]. What degree of vacuum the glove box can reach not only depends on quite different manufacturing costs, but also affects the effectiveness of vacuum preservation of sensitive items, and vacuum level is also an important key to effectively remove moisture and oxygen in the box. Usually, a high efficiency glove box must be the high vacuum box (transfer box) [10–12].

For high vacuum deoxygenation and water removal glove box, the oxygen content and water content in the box must usually reach the ppm level (i.e., up to 10<sup>−</sup><sup>4</sup> %) of the micro-oxygen and micro-water concentration [13–15]. However, it is difficult to achieve effectively such micro-oxygen and micro-water concentration. Usually, the glove box must be vacuumed, and then the box is filled with an inert gas (such as nitrogen or helium), and this type of evacuation and nitrogen (or helium) filling operation procedure must be performed multiple times. Thus, the deconcentration of water and oxygen can be accelerated, and the concentration of water and oxygen in the ppm range can be achieved more efficiently. In addition, the micro-water analyzer and micro-oxygen analyzer that detect ppm levels of water and oxygen concentration also have a detection range of ppm (e.g., 0–1000 ppm). To ensure the normal use of such analyzers, the input of water and oxygen concentration values of this type of analyzer should also fall within its detection range to avoid damage to the analyzer. If water and oxygen of high concentration (e.g., percentage level) are input for a long time, these analyzers will be easily damaged and cannot be used any more.

On the other hand, the transfer box of the glove box system is mainly for the user to put the materials into the glove box body or take out the materials from the box body. However, when the glove box body has reached the micro-oxygen and microwater concentration, the material should be placed through the transfer box. The material must be first put into the transfer box, and then the transfer box is sealed and evacuated to make the water and oxygen content similar to the glove box. But the transfer box in the vacuum state will suck, making it difficult to operate, so the inert gas in the glove box must be introduced into the transfer box so that the pressure is balanced, and then the isolation glove can be used to place the material into the box. On the contrary, if the material is to be taken out of the glove box and placed back in the transfer box (to be removed from the transfer box), the transfer box must be evacuated first, and then the inert gas must be introduced from the glove box to balance the pressure.

Although the glove boxes commonly used in various industries currently have a control system, they mainly monitor the vacuum system and the gas circulation exchange system through PLC in a semi-automatic manner [15, 16], and they must be operated by people by starting and setting the parameter of the vacuum system or gas cycling exchange system, so that the vacuum operation or gas cycle cleaning operations of such a glove box often need to wait for the completion of the parameter setting for the next stage, which tends to cause inconvenience in use and operation; on the other hand, to analyze the effectiveness of the vacuum operation and cleaning operation of the glove box system, it is necessary to automatically record the required measurement information for processing during each operation. However, no major glove box system is currently available with automatic

**37**

user of the glove box to plan experiments or operations.

*The Design and Development of Control System for High Vacuum Deoxygenated...*

recording of measurement data. The PLC control system is also very inconvenient for the instantaneous recording and processing of measurement data. Therefore, the existing glove box systems do not provide users with efficiency analysis function

Furthermore, since the commonly used glove boxes are equipped with only one gas circulation exchange system, namely, a reaction tank capable of removing moisture and oxygen through the reagent [15, 16], when the chemical reagent in the reaction tank is used up, the operation must be stopped. After the cleaning reagent is replenished, replaced, or regenerated, the cleaning operation can then resume. The regeneration of the reaction tank is usually by burning a low concentration of hydrogen to restore the cleaning function of the reagent [16]. However, during the experiment or operation in the glove box, due to the slight leakage of the glove box or some gas generated by the experimental operation, the internal environment of the box will constantly change, so that the box must be continuously monitored and the cleaning operation must be performed when the concentration of water or oxygen exceeds the specified value [11, 17]. Therefore, when the glove box system is only configured with one reaction tank, except for the inconvenience of use and operation, it may easily lead to the problem of interruption of the experiment or the operation process, thereby affecting the quality and efficiency of the glove box. Although there are currently a small number of larger glove box systems that can be configured with two or more sets of reaction tanks [16], the configured redundant reaction tanks are usually only alternative replacement devices. When the reagent of the reaction tank is used up and the cleaning function is lost, it must be switched to other reaction tank to continue the cleaning work by manual operations. In addition, the operation of replacing or regenerating the reagent is performed on the reaction tanks that have lost the cleaning function. However, this operation method still cannot solve the problem of interruption in the experiment or operation process.

In view of the defect of experimental or operational interruptions in the use of the glove box, a cycling cleaning regeneration mechanism was proposed in this study [18, 19]. This mechanism is mainly to build two reaction tanks A and B. At the beginning, the reaction tank A performs the cleaning work, and the reaction tank B takes the rest. When the reaction tank A loses the cleaning function due to the use up of the reagent, it will automatically switch to the tank B to continue to work, and the regeneration operation of the tank A starts at the same time. When the tank B loses the cleaning function due to the use up of the reagent, it immediately switches to the operation of tank A and simultaneously performs the regeneration of tank B. Thus, the two reaction tanks are used alternately to perform the cycle cleaning and recycling operation, which can effectively solve the problem of interruption in the experiment or the operation process of the glove box system. Based on this cycle cleaning regeneration mechanism, this study developed an advanced control system for high vacuum deoxygenation and water removal glove box. Through the design of switching by the hardware pipeline and control by the monitoring software, the system provides transfer box cleaning, glove box cleaning, vacuum preservation, end of preservation, cycling cleaning regeneration, and other basic operating functions on the one hand, and each operating function can be automatically executed through the setting of the control parameters and the instrument parameters; on the other hand, during the execution of each operating function, the system can automatically record all the measured data and elastically display the vacuum test curves and the cleaning test curves. This can provide the user with the analysis of the efficiency of cleaning operation and vacuum operation of the glove box. The results of this kind of performance analysis can be used as a quality parameter to determine the quality of the glove box and the basis for the

*DOI: http://dx.doi.org/10.5772/intechopen.80423*

for vacuum operation and cleaning operation.

#### *The Design and Development of Control System for High Vacuum Deoxygenated... DOI: http://dx.doi.org/10.5772/intechopen.80423*

recording of measurement data. The PLC control system is also very inconvenient for the instantaneous recording and processing of measurement data. Therefore, the existing glove box systems do not provide users with efficiency analysis function for vacuum operation and cleaning operation.

Furthermore, since the commonly used glove boxes are equipped with only one gas circulation exchange system, namely, a reaction tank capable of removing moisture and oxygen through the reagent [15, 16], when the chemical reagent in the reaction tank is used up, the operation must be stopped. After the cleaning reagent is replenished, replaced, or regenerated, the cleaning operation can then resume. The regeneration of the reaction tank is usually by burning a low concentration of hydrogen to restore the cleaning function of the reagent [16]. However, during the experiment or operation in the glove box, due to the slight leakage of the glove box or some gas generated by the experimental operation, the internal environment of the box will constantly change, so that the box must be continuously monitored and the cleaning operation must be performed when the concentration of water or oxygen exceeds the specified value [11, 17]. Therefore, when the glove box system is only configured with one reaction tank, except for the inconvenience of use and operation, it may easily lead to the problem of interruption of the experiment or the operation process, thereby affecting the quality and efficiency of the glove box. Although there are currently a small number of larger glove box systems that can be configured with two or more sets of reaction tanks [16], the configured redundant reaction tanks are usually only alternative replacement devices. When the reagent of the reaction tank is used up and the cleaning function is lost, it must be switched to other reaction tank to continue the cleaning work by manual operations. In addition, the operation of replacing or regenerating the reagent is performed on the reaction tanks that have lost the cleaning function. However, this operation method still cannot solve the problem of interruption in the experiment or operation process.

In view of the defect of experimental or operational interruptions in the use of the glove box, a cycling cleaning regeneration mechanism was proposed in this study [18, 19]. This mechanism is mainly to build two reaction tanks A and B. At the beginning, the reaction tank A performs the cleaning work, and the reaction tank B takes the rest. When the reaction tank A loses the cleaning function due to the use up of the reagent, it will automatically switch to the tank B to continue to work, and the regeneration operation of the tank A starts at the same time. When the tank B loses the cleaning function due to the use up of the reagent, it immediately switches to the operation of tank A and simultaneously performs the regeneration of tank B. Thus, the two reaction tanks are used alternately to perform the cycle cleaning and recycling operation, which can effectively solve the problem of interruption in the experiment or the operation process of the glove box system.

Based on this cycle cleaning regeneration mechanism, this study developed an advanced control system for high vacuum deoxygenation and water removal glove box. Through the design of switching by the hardware pipeline and control by the monitoring software, the system provides transfer box cleaning, glove box cleaning, vacuum preservation, end of preservation, cycling cleaning regeneration, and other basic operating functions on the one hand, and each operating function can be automatically executed through the setting of the control parameters and the instrument parameters; on the other hand, during the execution of each operating function, the system can automatically record all the measured data and elastically display the vacuum test curves and the cleaning test curves. This can provide the user with the analysis of the efficiency of cleaning operation and vacuum operation of the glove box. The results of this kind of performance analysis can be used as a quality parameter to determine the quality of the glove box and the basis for the user of the glove box to plan experiments or operations.

*Applied Modern Control*

10<sup>−</sup><sup>4</sup>

The glove box is mainly composed of a glove box body, a vacuum system, a gas circulation exchange system, and a control system [5]. The glove box body is provided with an antechamber (also called a transfer box) and an isolation glove. Usually a viewing window is arranged on the front of the body, so that the operator can clearly observe the operating conditions inside the box, and the operation process can be intuitively displayed in front of the operator. The glove box facilitates operators with isolation gloves in an anhydrous, oxygen free, and vacuum environment. Basically, the vacuum glove box is an important application of vacuum technology [6–9]. What degree of vacuum the glove box can reach not only depends on quite different manufacturing costs, but also affects the effectiveness of vacuum preservation of sensitive items, and vacuum level is also an important key to effectively remove moisture and oxygen in the box. Usually, a high efficiency glove box

For high vacuum deoxygenation and water removal glove box, the oxygen content and water content in the box must usually reach the ppm level (i.e., up to

water and oxygen concentration also have a detection range of ppm

%) of the micro-oxygen and micro-water concentration [13–15]. However, it is difficult to achieve effectively such micro-oxygen and micro-water concentration. Usually, the glove box must be vacuumed, and then the box is filled with an inert gas (such as nitrogen or helium), and this type of evacuation and nitrogen (or helium) filling operation procedure must be performed multiple times. Thus, the deconcentration of water and oxygen can be accelerated, and the concentration of water and oxygen in the ppm range can be achieved more efficiently. In addition, the micro-water analyzer and micro-oxygen analyzer that detect ppm levels of

(e.g., 0–1000 ppm). To ensure the normal use of such analyzers, the input of water and oxygen concentration values of this type of analyzer should also fall within its detection range to avoid damage to the analyzer. If water and oxygen of high concentration (e.g., percentage level) are input for a long time, these analyzers will

On the other hand, the transfer box of the glove box system is mainly for the user to put the materials into the glove box body or take out the materials from the box body. However, when the glove box body has reached the micro-oxygen and microwater concentration, the material should be placed through the transfer box. The material must be first put into the transfer box, and then the transfer box is sealed and evacuated to make the water and oxygen content similar to the glove box. But the transfer box in the vacuum state will suck, making it difficult to operate, so the inert gas in the glove box must be introduced into the transfer box so that the pressure is balanced, and then the isolation glove can be used to place the material into the box. On the contrary, if the material is to be taken out of the glove box and placed back in the transfer box (to be removed from the transfer box), the transfer box must be evacuated first, and then the inert gas must be introduced from the

Although the glove boxes commonly used in various industries currently have a control system, they mainly monitor the vacuum system and the gas circulation exchange system through PLC in a semi-automatic manner [15, 16], and they must be operated by people by starting and setting the parameter of the vacuum system or gas cycling exchange system, so that the vacuum operation or gas cycle cleaning operations of such a glove box often need to wait for the completion of the parameter setting for the next stage, which tends to cause inconvenience in use and operation; on the other hand, to analyze the effectiveness of the vacuum operation and cleaning operation of the glove box system, it is necessary to automatically record the required measurement information for processing during each operation. However, no major glove box system is currently available with automatic

must be the high vacuum box (transfer box) [10–12].

be easily damaged and cannot be used any more.

glove box to balance the pressure.

**36**

To verify the function of the deoxygenation and dehydration of the glove box system developed, a hydrogen storage and cleaning mechanism was built in the glove box of this study to serve as a specific application of the glove box system, and the application functions of hydrogen storage and hydrogen cell cleaning were provided in the monitoring software. This kind of design mainly considers that hydrogen is the main fuel for hydrogen vehicles in industry, and usually needs to be stored in a hydrogen storage cell. The hydrogen storage operation of the hydrogen storage cell must be in an oxygen free environment, including prior vacuumizing to remove the air in the hydrogen storage cell, then filling the hydrogen storage cell with hydrogen in an oxygen free environment, so as to avoid the danger of explosion due to the combination of hydrogen and oxygen. Therefore, we use this micro-oxygen and microwater glove box as the operating environment of the hydrogen storage cell, providing operators with the ability to safely and smoothly perform the operation of evacuating the hydrogen storage cell and storing hydrogen for hydrogen storage purpose.

### **2. System design**

The system architecture of the high vacuum deoxygenation and water removal glove box control system developed in this study is shown in **Figure 1**. In the figure, the solid blue line represents the gas flow path, the green dotted line represents the

**39**

*The Design and Development of Control System for High Vacuum Deoxygenated...*

detection signal transmission path, and the red dotted line represents the control

The functions of each component unit in **Figure 1** are described as follows:

1.Glove box body: a box body of oxygen free and water free environment that provides a high vacuum for the user to perform material handling or storage. The body contains a set of hydrogen cell units that can store hydrogen or remove

2.Transfer box: it is used for users to put the material into the glove box or remove

3.Body temperature detection unit: the temperature sensor used to detect the temperature of the glove box body, and the detected temperature can be transmit-

4.Primary humidity detection unit: it can detect the humidity of the body in the atmospheric state, and it is a humidity sensor of the percentage range detection.

5.Primary oxygen detection unit: it can detect the oxygen concentration of the body in the atmospheric state, and it is an oxygen sensor of the percentage range

6.Partial pressure control unit: it is a control module used to introduce the inert gas from the body into the transfer box to achieve a pressure balance.

7.Vacuum pressure detection unit: the pressure sensor used to detect the gas pressure (i.e., vacuum level) in the body, and the detected pressure can be transmit-

8.Pure gas unit: source of purified inert gas supply for high pressure nitrogen or

hydrogen provided by the pure gas unit and the hydrogen supply unit are of very high pressure and cannot be directly used, the decompression unit must be used to decompress the high pressure gas before the gas enters the glove box system.

9.Gas decompression unit: since the high pressure inert gas or high pressure

10.Gas supplement control unit: the control module to receive the command from the monitoring host, which is used to supplement the decompressed

11.Vacuum removal control unit: the vacuum pump controlled by the monitoring host to remove the gas in the body or the transfer box and bring it to a

12.Micro-oxygen detection unit: a micro-oxygen detector that can detect ppm oxygen concentration. The detected oxygen concentration can be transmitted

13.Micro-water detection unit: a micro-water detector that can detect ppm level moisture concentration, and the detected water concentration can be trans-

pure gas to the body or supply it to the regeneration reaction tank.

*DOI: http://dx.doi.org/10.5772/intechopen.80423*

the material from the box.

ted to the monitoring host.

ted to the monitoring host.

signal line.

hydrogen.

detection.

helium.

vacuum state.

to the monitoring host.

mitted to the monitoring host.

#### **Figure 1.**

*Architecture of the glove box control system.*

*The Design and Development of Control System for High Vacuum Deoxygenated... DOI: http://dx.doi.org/10.5772/intechopen.80423*

detection signal transmission path, and the red dotted line represents the control signal line.

The functions of each component unit in **Figure 1** are described as follows:


*Applied Modern Control*

**2. System design**

To verify the function of the deoxygenation and dehydration of the glove box system developed, a hydrogen storage and cleaning mechanism was built in the glove box of this study to serve as a specific application of the glove box system, and the application functions of hydrogen storage and hydrogen cell cleaning were provided in the monitoring software. This kind of design mainly considers that hydrogen is the main fuel for hydrogen vehicles in industry, and usually needs to be stored in a hydrogen storage cell. The hydrogen storage operation of the hydrogen storage cell must be in an oxygen free environment, including prior vacuumizing to remove the air in the hydrogen storage cell, then filling the hydrogen storage cell with hydrogen in an oxygen free environment, so as to avoid the danger of explosion due to the combination of hydrogen and oxygen. Therefore, we use this micro-oxygen and microwater glove box as the operating environment of the hydrogen storage cell, providing operators with the ability to safely and smoothly perform the operation of evacuating

the hydrogen storage cell and storing hydrogen for hydrogen storage purpose.

The system architecture of the high vacuum deoxygenation and water removal glove box control system developed in this study is shown in **Figure 1**. In the figure, the solid blue line represents the gas flow path, the green dotted line represents the

**38**

**Figure 1.**

*Architecture of the glove box control system.*

#### *Applied Modern Control*


**41**

**Figure 2.**

*Hardware design P&ID diagram.*

*The Design and Development of Control System for High Vacuum Deoxygenated...*

25.Signal acquiring control unit: the Advantech USB-4711A multi-function data acquisition card [20] is used as a signal capture unit to convert the analog and digital signals captured by each sensor into digital data, which are then transmitted to the monitoring host for processing. In addition, the Advantech USB-4750 digital output control card [21] is used as a signal control unit to transmit the control signals generated by the monitoring host to each control

26.Monitoring host: it is a personal computer that can execute LabVIEW monitoring software and configures a touch screen as the user interface for the user to operate the system. The LabVIEW monitoring software provides users with various manipulation functions, and at the same time, it can automatically record the relevant data of various operations, for drawing and flexible

In the hardware design part, the P&ID (Process and Instrument Diagram) of the glove box control system is shown in **Figure 2**, where R-A and R-B represent the reaction tanks A and B, respectively; PI is the pressure indicator; PT, TT, HT,

*DOI: http://dx.doi.org/10.5772/intechopen.80423*

unit to control each valve switch.

display of various test curves.

*The Design and Development of Control System for High Vacuum Deoxygenated... DOI: http://dx.doi.org/10.5772/intechopen.80423*


In the hardware design part, the P&ID (Process and Instrument Diagram) of the glove box control system is shown in **Figure 2**, where R-A and R-B represent the reaction tanks A and B, respectively; PI is the pressure indicator; PT, TT, HT,

**Figure 2.** *Hardware design P&ID diagram.*

*Applied Modern Control*

oxygen.

and B.

ing host.

(catalyst) in the reaction tank.

hydrogen storage operation.

tion is performed.

tion tanks A and B.

14.Sampling control unit: the sampling box controlled by the monitoring host may be used only for capturing gas in the body during the cycling cleaning operation for the micro-oxygen detector and the micro-water detector to detect the water and oxygen concentration so as to prevent such an expensive detector from being damaged by excessive concentrations of moisture or

15.Cleaning regeneration unit: it consists of two reaction tanks that can carry out cleaning and regeneration functions. The cleaning function is to use the cleaning reagent to adsorb oxygen and water vapor that flows through the reaction tank. When the regeneration function is performed, low concentra-

16.Cycle control unit: it can be controlled to perform the cycling cleaning and the cycling regeneration operations alternately of the two reaction tanks A

17.A/B channel switching unit: it is used to switch the pipeline channel of reac-

18.Preheated burning regeneration control unit: it can receive the command from the monitoring host for heating of the reaction tanks A or B and to remove oxygen and water vapor adsorbed on the chemicals by burning low

19.Regeneration temperature detection unit: a temperature sensor that can be used to detect the heating temperature when the reaction tank performs a regeneration function. Each of the reaction tanks A and B has a temperature sensor. The detected heating temperature can be transmitted to the monitor-

20.Low concentration hydrogen supply unit: it provides 3% concentration of hydrogen and nitrogen mixed gas, and this low concentration of hydrogen can be burnt to activate the regeneration function of the cleaning reagent

21.Outlet switching unit: it is used to switch the flow path of the gas from the outlet of the reaction tank A or reaction tank B. When the reaction tank performs the cleaning operation, the outlet gas flows back to the glove box body, and the outlet gas is directly discharged when the regeneration opera-

22.Hydrogen supply unit: the hydrogen source module for hydrogen supply.

24.Hydrogen index detection unit: it is used to detect the hydrogen index in the box body and send it to the monitoring host. If the hydrogen index is too high, there is a danger of explosion. The monitoring host will stop the

23.Hydrogen storage control unit: the module unit is controlled by the monitoring host to store and release hydrogen into the hydrogen storage cell.

concentration hydrogen to achieve regeneration.

tion hydrogen gas is introduced to regenerate its cleaning agent.

**40**

O2T, and H2T indicate the detecting transducers of pressure, temperature, humidity, oxygen concentration, and hydrogen concentration; PIC, TIC, HIC, and H2IC, respectively, indicate the indication controllers of pressure, temperature, humidity, and hydrogen concentration. All indication controllers inside the box can send signals to the USB-4711A adapter card and be processed by the monitoring host. TA and HA indicate the temperature and humidity actuators, respectively, and PVC indicates the pressure regulating valve. Each EV is an electric valve that can be controlled by the monitoring host via a USB-4750 adapter card.

To obtain high measurement quality of signals, such as temperature, pressure, water/oxygen concentration, and hydrogen concentration, etc., in terms of hardware, this study removed excessively large external signals through the masking technique and filtered the high frequency noises by virtue of short circuit filter capacitance. While in terms of software, this study employed the method of mean value to remove the influence of a small number of surges. For example, in a sampling period, the system captured 10 signals, calculated their mean, and used it as the measurement value, so that accurate and precise measurement signals can be obtained.

The glove box control system entity developed in this study is shown in **Figure 3**. We installed an internal pressure balanced end cap on the outside of the isolation gloves of the glove box to seal the isolation gloves inside the end cap. During vacuum operation of the box body, vacuum is applied together with the inside of the end cap to maintain the pressure balance. This avoids the pressure difference between the inside and the outside of the body, which may cause the isolation glove to inflate or even burst.

#### **3. Software development**

In this study, the automatic monitoring software for this glove box control system was developed using the LabVIEW graphical language [22–24]. The module hierarchy is shown in **Figure 4**. The monitoring software includes three major functional modules of parameter settings, program monitoring, and curve drawing. The parameter setting module can provide control parameter setting and instrument parameter setting functions. The program monitoring module can provide six

**43**

**3.2 Real-time control process**

*The Design and Development of Control System for High Vacuum Deoxygenated...*

functions of monitoring, including transfer box cleaning, box body cleaning, cycling cleaning and regeneration, vacuum preservation, hydrogen storage tank cleaning, and hydrogen storage. And the curve drawing module can perform the function of

**Figure 5** shows the initial screen of the glove box automatic monitoring system designed by LabVIEW. Click the "Parameter Setting" button on the upper left of the screen to set the control parameters and instrument parameters. The seven green buttons on the right side of the screen are for the user to perform seven monitoring functions such as transfer box cleaning, box body cleaning, cycling cleaning and regeneration, vacuum preservation, end preservation, hydrogen storage tank cleaning, and hydrogen storage. Click the "Curve Drawing" button at the bottom left to draw and display the clear test curve and vacuum test curve.

Before executing various control functions, the user can first set various control parameters and instrument parameters. The setting screen of control parameters is shown in **Figure 6(a)**. There are 16 control parameters that can be set in this screen. The control parameters set by users include the vacuum pressures of box body, transfer box and hydrogen cell, the set temperature of reaction tanks for cycling regeneration, the concentrations of water and oxygen set to stop cycling cleaning, and the times T1–T9 set to control various operation functions. The instrument parameter setting can mainly provide the system designers to use the sensing instrument based on actual conditions to establish the software and hardware interface of the system, that is, the detection range of the sensor and the signal level relationship of the corresponding adapter card pin. The instrument parameter setting screen is as shown in **Figure 6(b)**. In the setting of instrument parameters, nine sensing ranges (for the regenerative temperature of reaction tanks A and B, the internal temperature of box body, the primary humidity and primary oxygen of box body, the concentrations of micro-water and micro-water, the vacuum pressure of vacuum pump, and the hydrogen explosion index) are given, and the corresponding signal levels of analog input pin AI0–AI8 are selected.

After completing the control parameter setting, the user can perform various operation functions such as transfer box cleaning, box body cleaning, cycling cleaning regeneration, vacuum preservation, end of preservation, storing hydrogen and hydrogen cell cleaning, etc. The state transition diagram as shown in **Figure 7**

drawing and displaying the cleaning test curve and the vacuum test curve.

**3.1 Control parameter and instrument parameter setting**

*DOI: http://dx.doi.org/10.5772/intechopen.80423*

*The hierarchy of automatic monitoring software modules.*

**Figure 4.**

*The Design and Development of Control System for High Vacuum Deoxygenated... DOI: http://dx.doi.org/10.5772/intechopen.80423*

**Figure 4.** *The hierarchy of automatic monitoring software modules.*

*Applied Modern Control*

O2T, and H2T indicate the detecting transducers of pressure, temperature, humidity, oxygen concentration, and hydrogen concentration; PIC, TIC, HIC, and H2IC, respectively, indicate the indication controllers of pressure, temperature, humidity, and hydrogen concentration. All indication controllers inside the box can send signals to the USB-4711A adapter card and be processed by the monitoring host. TA and HA indicate the temperature and humidity actuators, respectively, and PVC indicates the pressure regulating valve. Each EV is an electric valve that can be

To obtain high measurement quality of signals, such as temperature, pressure, water/oxygen concentration, and hydrogen concentration, etc., in terms of hardware, this study removed excessively large external signals through the masking technique and filtered the high frequency noises by virtue of short circuit filter capacitance. While in terms of software, this study employed the method of mean value to remove the influence of a small number of surges. For example, in a sampling period, the system captured 10 signals, calculated their mean, and used it as the measurement value, so that accurate and precise measurement signals can be

The glove box control system entity developed in this study is shown in **Figure 3**. We installed an internal pressure balanced end cap on the outside of the isolation gloves of the glove box to seal the isolation gloves inside the end cap. During vacuum operation of the box body, vacuum is applied together with the inside of the end cap to maintain the pressure balance. This avoids the pressure difference between the inside and the outside

controlled by the monitoring host via a USB-4750 adapter card.

of the body, which may cause the isolation glove to inflate or even burst.

In this study, the automatic monitoring software for this glove box control system was developed using the LabVIEW graphical language [22–24]. The module hierarchy is shown in **Figure 4**. The monitoring software includes three major functional modules of parameter settings, program monitoring, and curve drawing. The parameter setting module can provide control parameter setting and instrument parameter setting functions. The program monitoring module can provide six

**42**

obtained.

**Figure 3.**

*Glove box control system entity.*

**3. Software development**

functions of monitoring, including transfer box cleaning, box body cleaning, cycling cleaning and regeneration, vacuum preservation, hydrogen storage tank cleaning, and hydrogen storage. And the curve drawing module can perform the function of drawing and displaying the cleaning test curve and the vacuum test curve.

**Figure 5** shows the initial screen of the glove box automatic monitoring system designed by LabVIEW. Click the "Parameter Setting" button on the upper left of the screen to set the control parameters and instrument parameters. The seven green buttons on the right side of the screen are for the user to perform seven monitoring functions such as transfer box cleaning, box body cleaning, cycling cleaning and regeneration, vacuum preservation, end preservation, hydrogen storage tank cleaning, and hydrogen storage. Click the "Curve Drawing" button at the bottom left to draw and display the clear test curve and vacuum test curve.

#### **3.1 Control parameter and instrument parameter setting**

Before executing various control functions, the user can first set various control parameters and instrument parameters. The setting screen of control parameters is shown in **Figure 6(a)**. There are 16 control parameters that can be set in this screen. The control parameters set by users include the vacuum pressures of box body, transfer box and hydrogen cell, the set temperature of reaction tanks for cycling regeneration, the concentrations of water and oxygen set to stop cycling cleaning, and the times T1–T9 set to control various operation functions.

The instrument parameter setting can mainly provide the system designers to use the sensing instrument based on actual conditions to establish the software and hardware interface of the system, that is, the detection range of the sensor and the signal level relationship of the corresponding adapter card pin. The instrument parameter setting screen is as shown in **Figure 6(b)**. In the setting of instrument parameters, nine sensing ranges (for the regenerative temperature of reaction tanks A and B, the internal temperature of box body, the primary humidity and primary oxygen of box body, the concentrations of micro-water and micro-water, the vacuum pressure of vacuum pump, and the hydrogen explosion index) are given, and the corresponding signal levels of analog input pin AI0–AI8 are selected.

#### **3.2 Real-time control process**

After completing the control parameter setting, the user can perform various operation functions such as transfer box cleaning, box body cleaning, cycling cleaning regeneration, vacuum preservation, end of preservation, storing hydrogen and hydrogen cell cleaning, etc. The state transition diagram as shown in **Figure 7**

#### **Figure 5.**

*Initial screen of the glove box automatic monitoring system.*

#### **Figure 6.**

*Parameter setting screens. (a) Control parameter setting and (b) instrument parameter setting.*

was used to represent the real-time control process for various operational functions of the glove box. The ellipses in the figure represent the states of the system, and the thick ellipses represent the final states. The double-circle ellipse represents the composite state, which is used to represent another state transition diagram. The line connecting the ellipses represents the transition of the state, and an event is attached next to each of the state transition connection line. This means that the transition of the system state is due to the occurrence of this event. In addition, a horizontal line can be added below the event. The action below the horizontal line is the action that accompanies the event, as shown below.

In the state transition diagram of **Figure 7**, when the system starts up, it enters a "Wait" state and enters the initial screen as shown in **Figure 5**, and the user can click "Transfer box clean", "Box body clean", "Cycling clean regen", "Vacuum preserve", "End preserve", "Hydrogen cell clean" or "Store Hydrogen" buttons to start the desired operation function. The system will enter the state corresponding

**45**

**Figure 7.**

cleaning work.

*The Design and Development of Control System for High Vacuum Deoxygenated...*

to this operation function and start to execute its state transition diagram. The operating function of "Cycling clean regen" corresponds to a double-loop composite state. Entering this state will transfer to perform the "Cycling clean regen" as shown in **Figure 12**. After each state change graph of each operation function in the state transition diagram of **Figure 7** is executed, it will return to the "Wait" state for the user to select the next operation function, or the user may click the "System end"

The transfer box cleaning operation removes moisture and oxygen from the transfer box of the glove box to a set vacuum level. When this operation is performed, the system first turns on the transfer box vacuum valve EV-5 and starts the vacuum pump P-1 to extract the air, and then enters the "*Transport box clean*" state, as shown in **Figure 8(a)**, which is the screen to execute the transfer box cleaning. In this screen, the red path represents the opening of the electric valve. Then it waits for T8 time, enters the state of "*Detect transfer box vacuum pressure*", starts measuring the vacuum pressure PT-1 (representing the transfer box pressure), and waits for the pressure to reach the set transfer box vacuum level (determined by the control parameters set in **Figure 6(a)**). When the vacuum pressure of the PT-1 is reached, it enters the "*Reach transfer box vacuum pressure*" state. At this time, the EV-5 shutoff action is started first, the T2 time is waited for, then the P-1 is closed, and then the "*Transfer box balance*" state is entered, this will open the transfer box balance valve EV-1 and time T1 will be counted, in order to introduce the inert gas in the box body into the transfer box, so that it can achieve pressure balance, thus completing the transfer box

*DOI: http://dx.doi.org/10.5772/intechopen.80423*

button to end the execution of this system.

*State transition diagram of the glove box real-time control process.*

**3.3 Transfer box cleaning and box body cleaning operations**

*The Design and Development of Control System for High Vacuum Deoxygenated... DOI: http://dx.doi.org/10.5772/intechopen.80423*

**Figure 7.** *State transition diagram of the glove box real-time control process.*

to this operation function and start to execute its state transition diagram. The operating function of "Cycling clean regen" corresponds to a double-loop composite state. Entering this state will transfer to perform the "Cycling clean regen" as shown in **Figure 12**. After each state change graph of each operation function in the state transition diagram of **Figure 7** is executed, it will return to the "Wait" state for the user to select the next operation function, or the user may click the "System end" button to end the execution of this system.

#### **3.3 Transfer box cleaning and box body cleaning operations**

The transfer box cleaning operation removes moisture and oxygen from the transfer box of the glove box to a set vacuum level. When this operation is performed, the system first turns on the transfer box vacuum valve EV-5 and starts the vacuum pump P-1 to extract the air, and then enters the "*Transport box clean*" state, as shown in **Figure 8(a)**, which is the screen to execute the transfer box cleaning. In this screen, the red path represents the opening of the electric valve. Then it waits for T8 time, enters the state of "*Detect transfer box vacuum pressure*", starts measuring the vacuum pressure PT-1 (representing the transfer box pressure), and waits for the pressure to reach the set transfer box vacuum level (determined by the control parameters set in **Figure 6(a)**). When the vacuum pressure of the PT-1 is reached, it enters the "*Reach transfer box vacuum pressure*" state. At this time, the EV-5 shutoff action is started first, the T2 time is waited for, then the P-1 is closed, and then the "*Transfer box balance*" state is entered, this will open the transfer box balance valve EV-1 and time T1 will be counted, in order to introduce the inert gas in the box body into the transfer box, so that it can achieve pressure balance, thus completing the transfer box cleaning work.

*Applied Modern Control*

**Figure 5.**

**Figure 6.**

*Initial screen of the glove box automatic monitoring system.*

**44**

was used to represent the real-time control process for various operational functions of the glove box. The ellipses in the figure represent the states of the system, and the thick ellipses represent the final states. The double-circle ellipse represents the composite state, which is used to represent another state transition diagram. The line connecting the ellipses represents the transition of the state, and an event is attached next to each of the state transition connection line. This means that the transition of the system state is due to the occurrence of this event. In addition, a horizontal line can be added below the event. The action below the horizontal line is

*Parameter setting screens. (a) Control parameter setting and (b) instrument parameter setting.*

In the state transition diagram of **Figure 7**, when the system starts up, it enters

a "Wait" state and enters the initial screen as shown in **Figure 5**, and the user can click "Transfer box clean", "Box body clean", "Cycling clean regen", "Vacuum preserve", "End preserve", "Hydrogen cell clean" or "Store Hydrogen" buttons to start the desired operation function. The system will enter the state corresponding

the action that accompanies the event, as shown below.

**Figure 8.**

*Execution screen of transfer box cleaning and box body cleaning. (a) Transfer box cleaning procedure and (b) body cleaning procedure.*

The box body cleaning operation is used to remove the moisture and oxygen of the glove box itself so that it reaches the set vacuum level. When this operation is performed, the system first opens the box vacuum valve EV-3 and activates the vacuum pump P-1, and simultaneously opens the inlet valves CV1A, CV2A, CV1B, and CV2B of the regeneration system, to eliminate the oxygen and moisture in the pipeline. Then it enters the "*Box body clean*" state, as shown in **Figure 8(b)**, which is the screen to perform the box body cleaning. Then it waits for the T8 time, enters the next state of "*Detect body vacuum pressure*", starts to measure the vacuum pressure PT-1 (representing the body pressure), and waits for the pressure to reach the set vacuum level of the box body. When the vacuum pressure of PT-1 is reached, it will enter the "*Reach body vacuum pressure*" state. At this time, the EV-3 shutdown will be started first, the time will be waited for T2, then the P-1 will be closed, then the "*Detect box positive pressure*" state is entered, the glove box air supply replenishment valve EV-2 will be started, and N2 gas will be added. When the positive pressure of the box body detected is sufficient, the system will enter the state of "*Body positive pressure adequate*" and close the EV-2 to stop the replenishment of the N2 gas source and completes the box body cleaning work. The system will check if the set number of body cleaning has been reached. If it has not, it will return to the "*Box body clean*" state and start the next round of box cleaning until the set number of box body cleaning has been completed. Then it will close the EV-3, CV1A, CV2A, CV1B, and CV2B valves and end the entire box body cleaning work.

#### **3.4 Vacuum preservation and end preservation operations**

The vacuum preservation operation is used to start the procedure for storing sensitive materials in the glove box. To perform a vacuum preservation operation, the user can activate the vacuum preservation control process after the material to be stored is placed in the glove box. The system first opens the box body vacuum valve EV-3 and activates the vacuum pump P-1 and simultaneously turns on the inlet valves of the regeneration system, CV1A, CV2A, CV1B, and CV2B to remove the oxygen and water in the pipeline, and then enters the "*Vacuum preserve*" state. The execution screen for vacuum preservation is similar to the body cleaning screen, as shown in **Figure 8(a)**, which is the screen to perform vacuum pumping. Then it waits for the T8 time, then enters the "*Detect body vacuum pressure*" state, and starts to measure the vacuum pressure PT-1 to wait for this pressure to reach the set box body vacuum level. Then it enters the "*Reach body vacuum pressure*" state. At this time, the EV-3 is turned off, and the T2 timing is started, when the vacuum

**47**

out the stored material.

**Figure 9.**

*preservation procedure.*

*The Design and Development of Control System for High Vacuum Deoxygenated...*

pump P-1 continues to run, waiting for the EV-3 to complete the closing action, as shown in **Figure 9(a)**, which is the screen where the EV-3 is turned off first, and the P-1 continues to run. When the T2 time expires, the system will close the valves of CV1A, CV2A, CV1B, and CV2B. At this time, the box body can maintain a vacuum

*Execution screen for vacuum preservation and end preservation. (a) Closing the vacuum valve of the box body at the vacuum preservation procedure and (b) supplementing the box body with positive pressure at the end* 

The end preservation operation is used to start the procedure for ending the material storage. When the user finishes the preservation operation, the system will first enter the "*End preserve*" state, open the glove box gas source replenishment valve EV-2, and start to input the N2 gas to supplement the pressure of the box body and detect the box positive pressure PS-1, as shown in **Figure 9(b)**, which is the screen to supplement the box body positive pressure. When the detected positive pressure in the box body is sufficient, the system will first turn off the EV-2, stop the replenishment of the N2 gas source, then enter the "*Transfer box balance*" state, and the transfer box balancing valve EV-1 will be turned on. The gas in the body is introduced into the transfer box to bring it into pressure balance, then the time is counted to perform T1 time, and then the system enters the "*Exit preserve*" state. At this point, the user can open the inside and outside doors of the transfer box to take

The hydrogen cell cleaning operation is used to remove the gas in the hydrogen storage cell to reach the set vacuum level, so as to facilitate the subsequent operation of storing and releasing hydrogen. When this operation is performed, the system will first turn on the hydrogen cell vacuum valve EV-4 and start the vacuum pump P-1 to enter the "*Hydrogen cell clean*" state, and start the T8 timing, as shown in **Figure 10(a)**, which is the hydrogen cell cleaning screen. After waiting for the T8 timer to complete, the system will enter the "*Detect cell vacuum pressure*" state. At this time, the pressure PT-1 (representing the vacuum pressure of the hydrogen cell) will be measured. It is necessary to wait for the pressure to reach the set vacuum level. When the pressure of PT-1 reaches the set vacuum level, the system enters the "*Reach cell vacuum pressure*" state. At this time, the closing action of EV-4 is started first, and the time is waited for T2 time. When the time of T2 is reached,

Hydrogen storage operation can store and release hydrogen in the hydrogen cell in the glove box. When this operation is performed, the system first enters the "*Store* 

environment and the material storage operation can begin.

**3.5 Hydrogen cell cleaning and hydrogen storage operations**

the hydrogen cell cleaning operation can be ended.

*DOI: http://dx.doi.org/10.5772/intechopen.80423*

*The Design and Development of Control System for High Vacuum Deoxygenated... DOI: http://dx.doi.org/10.5772/intechopen.80423*

**Figure 9.**

*Applied Modern Control*

**Figure 8.**

*body cleaning procedure.*

*Execution screen of transfer box cleaning and box body cleaning. (a) Transfer box cleaning procedure and (b)* 

The box body cleaning operation is used to remove the moisture and oxygen of the glove box itself so that it reaches the set vacuum level. When this operation is performed, the system first opens the box vacuum valve EV-3 and activates the vacuum pump P-1, and simultaneously opens the inlet valves CV1A, CV2A, CV1B, and CV2B of the regeneration system, to eliminate the oxygen and moisture in the pipeline. Then it enters the "*Box body clean*" state, as shown in **Figure 8(b)**, which is the screen to perform the box body cleaning. Then it waits for the T8 time, enters the next state of "*Detect body vacuum pressure*", starts to measure the vacuum pressure PT-1 (representing the body pressure), and waits for the pressure to reach the set vacuum level of the box body. When the vacuum pressure of PT-1 is reached, it will enter the "*Reach body vacuum pressure*" state. At this time, the EV-3 shutdown will be started first, the time will be waited for T2, then the P-1 will be closed, then the "*Detect box positive pressure*" state is entered, the glove box air supply replenishment valve EV-2 will be started, and N2 gas will be added. When the positive pressure of the box body detected is sufficient, the system will enter the state of "*Body positive pressure adequate*" and close the EV-2 to stop the replenishment of the N2 gas source and completes the box body cleaning work. The system will check if the set number of body cleaning has been reached. If it has not, it will return to the "*Box body clean*" state and start the next round of box cleaning until the set number of box body cleaning has been completed. Then it will close the EV-3, CV1A, CV2A,

CV1B, and CV2B valves and end the entire box body cleaning work.

The vacuum preservation operation is used to start the procedure for storing sensitive materials in the glove box. To perform a vacuum preservation operation, the user can activate the vacuum preservation control process after the material to be stored is placed in the glove box. The system first opens the box body vacuum valve EV-3 and activates the vacuum pump P-1 and simultaneously turns on the inlet valves of the regeneration system, CV1A, CV2A, CV1B, and CV2B to remove the oxygen and water in the pipeline, and then enters the "*Vacuum preserve*" state. The execution screen for vacuum preservation is similar to the body cleaning screen, as shown in **Figure 8(a)**, which is the screen to perform vacuum pumping. Then it waits for the T8 time, then enters the "*Detect body vacuum pressure*" state, and starts to measure the vacuum pressure PT-1 to wait for this pressure to reach the set box body vacuum level. Then it enters the "*Reach body vacuum pressure*" state. At this time, the EV-3 is turned off, and the T2 timing is started, when the vacuum

**3.4 Vacuum preservation and end preservation operations**

**46**

*Execution screen for vacuum preservation and end preservation. (a) Closing the vacuum valve of the box body at the vacuum preservation procedure and (b) supplementing the box body with positive pressure at the end preservation procedure.*

pump P-1 continues to run, waiting for the EV-3 to complete the closing action, as shown in **Figure 9(a)**, which is the screen where the EV-3 is turned off first, and the P-1 continues to run. When the T2 time expires, the system will close the valves of CV1A, CV2A, CV1B, and CV2B. At this time, the box body can maintain a vacuum environment and the material storage operation can begin.

The end preservation operation is used to start the procedure for ending the material storage. When the user finishes the preservation operation, the system will first enter the "*End preserve*" state, open the glove box gas source replenishment valve EV-2, and start to input the N2 gas to supplement the pressure of the box body and detect the box positive pressure PS-1, as shown in **Figure 9(b)**, which is the screen to supplement the box body positive pressure. When the detected positive pressure in the box body is sufficient, the system will first turn off the EV-2, stop the replenishment of the N2 gas source, then enter the "*Transfer box balance*" state, and the transfer box balancing valve EV-1 will be turned on. The gas in the body is introduced into the transfer box to bring it into pressure balance, then the time is counted to perform T1 time, and then the system enters the "*Exit preserve*" state. At this point, the user can open the inside and outside doors of the transfer box to take out the stored material.

#### **3.5 Hydrogen cell cleaning and hydrogen storage operations**

The hydrogen cell cleaning operation is used to remove the gas in the hydrogen storage cell to reach the set vacuum level, so as to facilitate the subsequent operation of storing and releasing hydrogen. When this operation is performed, the system will first turn on the hydrogen cell vacuum valve EV-4 and start the vacuum pump P-1 to enter the "*Hydrogen cell clean*" state, and start the T8 timing, as shown in **Figure 10(a)**, which is the hydrogen cell cleaning screen. After waiting for the T8 timer to complete, the system will enter the "*Detect cell vacuum pressure*" state. At this time, the pressure PT-1 (representing the vacuum pressure of the hydrogen cell) will be measured. It is necessary to wait for the pressure to reach the set vacuum level. When the pressure of PT-1 reaches the set vacuum level, the system enters the "*Reach cell vacuum pressure*" state. At this time, the closing action of EV-4 is started first, and the time is waited for T2 time. When the time of T2 is reached, the hydrogen cell cleaning operation can be ended.

Hydrogen storage operation can store and release hydrogen in the hydrogen cell in the glove box. When this operation is performed, the system first enters the "*Store* 

**Figure 10.**

*Execution screen for hydrogen storage tank cleaning and hydrogen storage. (a) Hydrogen cell cleaning procedure and (b) hydrogen storage procedure.*

*hydrogen*" state, opens the hydrogen storage shut-off valve EV-11, and inputs hydrogen into the hydrogen cell of the glove box, as shown in **Figure 10(b)**, which is the initial screen for hydrogen storage. Then waits T9 time to finish the hydrogen storage operation and enters the "*Store complete*" state. In this process of hydrogen storage, if the hydrogen concentration is detected too high, the green "Hydrogen normal" light (as shown in **Figure 10(b)**) will be changed to flashing "Hydrogen alarm" red light, and the hydrogen storage operation will be immediately stopped.

#### **4. Cycling cleaning regeneration control process**

The glove box system contains two reaction tanks A and B. The cycling cleaning regeneration function can cycle between the two reaction tanks. When one reaction tank loses the cleaning capacity due to the run out of reagent, it can switch automatically to another reaction tank to work, and the reaction tank that losing the cleaning capacity performs the regeneration operation at the same time before joining the cycling work after they resume the cleaning capacity. In other words, this system can perform the cycle cleaning and cycle regeneration between reaction tanks A and B synchronously. The synchronization process is shown in **Figure 11**. In the figure, T3–T7 are the time parameters set by the user in **Figure 6(a)**, where T3 is the cumulative maximum working time of the reaction tanks (and the tanks can no longer work beyond this time), T4 is the time to regenerate the reagent by introducing 3% hydrogen, T5 is the time for the introduction of nitrogen to remove oxygen and water, T6 and T7 are the working time and the rest time, respectively, of the reaction tanks for each operation.

**Figure 12** shows the state transition diagram of the glove box cycling cleaning regeneration control process. The dashed circles (states) and the arrow lines represent the parts that can be executed synchronously. Reaction tank A is preset to be the tank for the first time execution by the system, and the reaction tank B is in standby state.

The control system first enters the "*A works*" state, executes the cycling cleaning procedure of reaction tank A, it opens the cycling inlet valve CV1A of tank A, the cycling outlet valve CV2A, the oxygen sampling valve EV-6, the humidity sampling valve EV-7, etc., opens the clean pump P-2, and starts the timing of T3 and T6 at the same time, as shown in **Figure 13(a)**, which is the screen for tank A to perform the cleaning. At this point, the clean pump P-2 starts to send the oxygen and moisture in the glove box continuously to the reaction tank A for adsorption, so as to achieve

**49**

**Figure 12.**

**Figure 11.**

*The Design and Development of Control System for High Vacuum Deoxygenated...*

the purpose of removing oxygen and moisture. When the T6 time is completed, the system enters the "*A rests*" state. At this time, the valves CV1A, CV2A, EV-6, EV-7, and the clean pump P-2 are all closed, and the T7 timing for rest starts, as shown in **Figure 13(b)**, which is the screen of the reaction tank A taking rest for the time of T7. When the T7 timing is completed, the system will enter the status of "*A clean* 

*once*". At this time, the system will check whether both the detected oxygen

*DOI: http://dx.doi.org/10.5772/intechopen.80423*

*Synchronization process flow of cycling cleaning regeneration.*

*State transition diagram of cycling cleaning regeneration control process.*

*The Design and Development of Control System for High Vacuum Deoxygenated... DOI: http://dx.doi.org/10.5772/intechopen.80423*

#### **Figure 11.**

*Applied Modern Control*

**Figure 10.**

*and (b) hydrogen storage procedure.*

*hydrogen*" state, opens the hydrogen storage shut-off valve EV-11, and inputs hydrogen into the hydrogen cell of the glove box, as shown in **Figure 10(b)**, which is the initial screen for hydrogen storage. Then waits T9 time to finish the hydrogen storage operation and enters the "*Store complete*" state. In this process of hydrogen storage, if the hydrogen concentration is detected too high, the green "Hydrogen normal" light (as shown in **Figure 10(b)**) will be changed to flashing "Hydrogen alarm" red light,

*Execution screen for hydrogen storage tank cleaning and hydrogen storage. (a) Hydrogen cell cleaning procedure* 

The glove box system contains two reaction tanks A and B. The cycling cleaning regeneration function can cycle between the two reaction tanks. When one reaction tank loses the cleaning capacity due to the run out of reagent, it can switch automatically to another reaction tank to work, and the reaction tank that losing the cleaning capacity performs the regeneration operation at the same time before joining the cycling work after they resume the cleaning capacity. In other words, this system can perform the cycle cleaning and cycle regeneration between reaction tanks A and B synchronously. The synchronization process is shown in **Figure 11**. In the figure, T3–T7 are the time parameters set by the user in **Figure 6(a)**, where T3 is the cumulative maximum working time of the reaction tanks (and the tanks can no longer work beyond this time), T4 is the time to regenerate the reagent by introducing 3% hydrogen, T5 is the time for the introduction of nitrogen to remove oxygen and water, T6 and T7 are the working time and the rest time, respectively, of

**Figure 12** shows the state transition diagram of the glove box cycling cleaning regeneration control process. The dashed circles (states) and the arrow lines represent the parts that can be executed synchronously. Reaction tank A is preset to be the tank for the first time execution by the system, and the reaction tank B is in

The control system first enters the "*A works*" state, executes the cycling cleaning procedure of reaction tank A, it opens the cycling inlet valve CV1A of tank A, the cycling outlet valve CV2A, the oxygen sampling valve EV-6, the humidity sampling valve EV-7, etc., opens the clean pump P-2, and starts the timing of T3 and T6 at the same time, as shown in **Figure 13(a)**, which is the screen for tank A to perform the cleaning. At this point, the clean pump P-2 starts to send the oxygen and moisture in the glove box continuously to the reaction tank A for adsorption, so as to achieve

and the hydrogen storage operation will be immediately stopped.

**4. Cycling cleaning regeneration control process**

the reaction tanks for each operation.

**48**

standby state.

*Synchronization process flow of cycling cleaning regeneration.*

#### **Figure 12.**

*State transition diagram of cycling cleaning regeneration control process.*

the purpose of removing oxygen and moisture. When the T6 time is completed, the system enters the "*A rests*" state. At this time, the valves CV1A, CV2A, EV-6, EV-7, and the clean pump P-2 are all closed, and the T7 timing for rest starts, as shown in **Figure 13(b)**, which is the screen of the reaction tank A taking rest for the time of T7.

When the T7 timing is completed, the system will enter the status of "*A clean once*". At this time, the system will check whether both the detected oxygen

**Figure 13.**

*Screen for reaction tank A to perform cleaning operation. (a) Cleaning operation of reaction tank A and (b) reaction tank A taking rest for T7 time.*

concentration OT-2 and the moisture concentration HT-2 have reached the set concentrations. If yes, the system will enter "*A clean complete*" state and automatically return to the "*Wait*" state of **Figure 6**. If OT-2 and HT-2 cannot reach the set concentrations at the same time, the system will check whether the cumulative working time T3 has been reached. If the T3 time has not yet been reached, it will return to the "*A works*" state and continue to perform the cycling procedure of tank A working first for T6 time and then resting for T7 time.

If T3 time has been reached and OT-2 and HT-2 have not yet reached the set concentrations, it means that the reaction tank A has lost the cleaning ability due to the run out of the reagent, and the system will enter the state of "*Switch to B*". This state will start the reaction tank A regeneration function first, and then move to the "*B works*" state for reaction tank B to perform the cleaning work, while tank A performs the regeneration process at the same time, as shown in **Figure 13(a)**, which is the screen for tank B to work and tank A in regeneration. When entering the "*B works*" state, the system will first open the CV1B, CV2B, EV-6, and EV-7 valves and the clean pump P-2 and start the timing of T6 and T3. Then it performs the cycling cleaning procedure of working for T6 time and then resting for T7. This control process is similar to the monitoring process when tank A is working.

When tank B performs the cleaning operation, tank A executes the regeneration process synchronously, and the state transition diagram thereof is shown in the rightmost blue dotted state area of **Figure 12**. Reaction tank A first enters the "*A regenerate heating*" state. The system controls the heaters to heat the reaction tank A [25–27], and detects whether the temperature TT-A has reached the set regeneration temperature. When the reaction tank A's temperature TT-A reaches the set temperature, the system will enter the "*A reduce and exhaust*" state. At this time, the 3% hydrogen inlet valve EV-8A and exhaust valve EV-9A will be turned on for 3% hydrogen to enter the reaction tank A to regenerate its cleaning agent, for the reduced waste gas to be discharged via the EV-9A valve, and the timing of T4 starts at the same time, as shown in **Figure 14(a)**, which is the screen for tank B to perform the cleaning work (which is now in the T7 rest phase) and for tank A to enter 3% hydrogen to regenerate its cleaning reagent.

When the T4 time expires, the system will enter the "*A inputs N2 and exhaust*" state. At this time, the 3% hydrogen gas inlet valve EV-8A will be closed first, but the cleaning valve EV-10A will be opened to introduce dry N2 gas in the gas source, so that the oxygen and moisture in the reaction tank A are taken away and discharged through the outlet valve EV-9A. And the timing of T5 starts, as shown in

**51**

*The Design and Development of Control System for High Vacuum Deoxygenated...*

**Figure 14(b)**, which is the screen where tank B operates and tank A introduces dry

*Screen of tank B performing cleaning operation and tank A in regeneration. (a) Tank B resting for T7 and tank A entering 3% hydrogen to regenerate the reagent and (b) tank B operating and tank A introducing dry gas to* 

The system automatically records the detected input signals for archiving during the execution of various glove box operation functions. The detected input signals are shown in the screen of **Figure 5**. By using the internal temperature TT-1, the primary humidity HT-1, the primary oxygen concentration OT-1, the vacuum pressure PT-1, and the PT-1 set value, the system can draw the vacuum test curves of the operation functions such as the transfer box cleaning, box body cleaning, and vacuum preservation, as shown in **Figure 15**, which is the elastic display screen of the vacuum test curve corresponding to the box body cleaning. **Figure 15(a)** shows the complete vacuum test curves. There are five switch buttons and five curve color setting boxes of "TT-1," "HT-1," "OT-1," "PT-1," and "PT-1 setup" on the left of the screen. Each switch button can be used to switch between "show/hide." Each color setting box enables the user to set the display color of the corresponding curve. At present, all five buttons are in the "show" state. Therefore, the above five test curves are displayed in the vacuum test graph in the middle of the screen. All the curves are normalized so that the display range is from 0 to 100%. From **Figure 15(a)**, we can see that the rate of decrease of water vapor concentration (72 → 2%) is higher than

The "show/hide" status of each switch button is set appropriately, allowing the user to analyze the relationship between various curves. If you want to analyze the vacuum pressure and the change of the primary water vapor concentration in

When the T5 time expires, the system enters the "*A regenerate complete*" state. At this time, EV-9A and EV-10A are turned off to end tank A regeneration operation. When tank B has several cycling of operation by working for T6 time and resting for T7 time several times and the detected oxygen and water concentration still fail to reach the set values, this means the cleaning agent has also been used up. At this time, as long as the T3 time has been reached, it will also start tank B regeneration and then enter the state of "*Switch to A*", to give the cleaning work back to tank A,

*DOI: http://dx.doi.org/10.5772/intechopen.80423*

gas to take away the oxygen and moisture.

**Figure 14.**

*take away oxygen and moisture.*

and tank B will perform the regeneration work.

**5.1 Vacuum test curve rendering analysis**

the rate of decline of oxygen concentration (26 → 2%).

**5. Curve rendering analysis**

*The Design and Development of Control System for High Vacuum Deoxygenated... DOI: http://dx.doi.org/10.5772/intechopen.80423*

**Figure 14.**

*Applied Modern Control*

**Figure 13.**

*reaction tank A taking rest for T7 time.*

concentration OT-2 and the moisture concentration HT-2 have reached the set concentrations. If yes, the system will enter "*A clean complete*" state and automatically return to the "*Wait*" state of **Figure 6**. If OT-2 and HT-2 cannot reach the set concentrations at the same time, the system will check whether the cumulative working time T3 has been reached. If the T3 time has not yet been reached, it will return to the "*A works*" state and continue to perform the cycling procedure of tank

*Screen for reaction tank A to perform cleaning operation. (a) Cleaning operation of reaction tank A and (b)* 

If T3 time has been reached and OT-2 and HT-2 have not yet reached the set concentrations, it means that the reaction tank A has lost the cleaning ability due to the run out of the reagent, and the system will enter the state of "*Switch to B*". This state will start the reaction tank A regeneration function first, and then move to the "*B works*" state for reaction tank B to perform the cleaning work, while tank A performs the regeneration process at the same time, as shown in **Figure 13(a)**, which is the screen for tank B to work and tank A in regeneration. When entering the "*B works*" state, the system will first open the CV1B, CV2B, EV-6, and EV-7 valves and the clean pump P-2 and start the timing of T6 and T3. Then it performs the cycling cleaning procedure of working for T6 time and then resting for T7. This control process is similar to the monitoring process when

When tank B performs the cleaning operation, tank A executes the regeneration process synchronously, and the state transition diagram thereof is shown in the rightmost blue dotted state area of **Figure 12**. Reaction tank A first enters the "*A regenerate heating*" state. The system controls the heaters to heat the reaction tank A [25–27], and detects whether the temperature TT-A has reached the set regeneration temperature. When the reaction tank A's temperature TT-A reaches the set temperature, the system will enter the "*A reduce and exhaust*" state. At this time, the 3% hydrogen inlet valve EV-8A and exhaust valve EV-9A will be turned on for 3% hydrogen to enter the reaction tank A to regenerate its cleaning agent, for the reduced waste gas to be discharged via the EV-9A valve, and the timing of T4 starts at the same time, as shown in **Figure 14(a)**, which is the screen for tank B to perform the cleaning work (which is now in the T7 rest phase) and for tank A to

When the T4 time expires, the system will enter the "*A inputs N2 and exhaust*" state. At this time, the 3% hydrogen gas inlet valve EV-8A will be closed first, but the cleaning valve EV-10A will be opened to introduce dry N2 gas in the gas source, so that the oxygen and moisture in the reaction tank A are taken away and discharged through the outlet valve EV-9A. And the timing of T5 starts, as shown in

A working first for T6 time and then resting for T7 time.

enter 3% hydrogen to regenerate its cleaning reagent.

**50**

tank A is working.

*Screen of tank B performing cleaning operation and tank A in regeneration. (a) Tank B resting for T7 and tank A entering 3% hydrogen to regenerate the reagent and (b) tank B operating and tank A introducing dry gas to take away oxygen and moisture.*

**Figure 14(b)**, which is the screen where tank B operates and tank A introduces dry gas to take away the oxygen and moisture.

When the T5 time expires, the system enters the "*A regenerate complete*" state. At this time, EV-9A and EV-10A are turned off to end tank A regeneration operation. When tank B has several cycling of operation by working for T6 time and resting for T7 time several times and the detected oxygen and water concentration still fail to reach the set values, this means the cleaning agent has also been used up. At this time, as long as the T3 time has been reached, it will also start tank B regeneration and then enter the state of "*Switch to A*", to give the cleaning work back to tank A, and tank B will perform the regeneration work.

#### **5. Curve rendering analysis**

#### **5.1 Vacuum test curve rendering analysis**

The system automatically records the detected input signals for archiving during the execution of various glove box operation functions. The detected input signals are shown in the screen of **Figure 5**. By using the internal temperature TT-1, the primary humidity HT-1, the primary oxygen concentration OT-1, the vacuum pressure PT-1, and the PT-1 set value, the system can draw the vacuum test curves of the operation functions such as the transfer box cleaning, box body cleaning, and vacuum preservation, as shown in **Figure 15**, which is the elastic display screen of the vacuum test curve corresponding to the box body cleaning. **Figure 15(a)** shows the complete vacuum test curves. There are five switch buttons and five curve color setting boxes of "TT-1," "HT-1," "OT-1," "PT-1," and "PT-1 setup" on the left of the screen. Each switch button can be used to switch between "show/hide." Each color setting box enables the user to set the display color of the corresponding curve. At present, all five buttons are in the "show" state. Therefore, the above five test curves are displayed in the vacuum test graph in the middle of the screen. All the curves are normalized so that the display range is from 0 to 100%. From **Figure 15(a)**, we can see that the rate of decrease of water vapor concentration (72 → 2%) is higher than the rate of decline of oxygen concentration (26 → 2%).

The "show/hide" status of each switch button is set appropriately, allowing the user to analyze the relationship between various curves. If you want to analyze the vacuum pressure and the change of the primary water vapor concentration in

**Figure 15.**

*Display screen of vacuum test curves of box body cleaning. (a) Complete vacuum test curve and (b) change relationship between PT-1 and HT-1.*

the box, you may click "TT-1" and "OT-1" switch buttons to set them to the "hide" status, and make the other three buttons "HT-1," "PT-1," and "PT-1 setup" maintain the "show" state, as shown in **Figure 15(b)**, which is the display of the relationship between the PT-1 and HT-1 curves. From this screen, it is observed that the green PT-1 pressure curve causes the significant drop of the blue HT-1 water vapor concentration during each descent (which represents the vacuumizing stage), and whether the PT-1 pressure curve can reach the gray PT-1 setup curve (i.e., reaching the setting value of the vacuum pressure) at each descent; and the pressure curve of PT-1 during the flat bottom is the nitrogen filling phase of the box body, when the change in the HT-1 water vapor concentration curve is less obvious. From the above analysis, we can see that the effect of the vacuumizing stage on the decline of water vapor concentration is obviously higher than that of the nitrogen filling stage.

#### **5.2 Cleaning test curve rendering analysis**

By using automatically recorded box temperature TT-1, regeneration temperature TT-A of the reaction tank A, regeneration temperature TT-B of the reaction tank B, the micro-water concentration HT-2 and micro-oxygen concentration OT-2 detected by the sampling tank, and the settings of HT-2 and OT-2, the system can draw the clean test curves of the cycling cleaning regeneration operation, as shown in **Figure 16**, which is the elastic display screen of the test curve of such cleaning. **Figure 16(a)** shows the complete cleaning test curves. There are eight switching buttons left side of the screen, such as "TT-1," "TT-A," "TT-B," "HT-2," "OT-2," "Regen Temp setup," "OT-2 setup," and "HT-2 setup" and eight corresponding curve color setting blocks. Each switching button can be used by the user to switch between show/hide, and each color setting box enables the user to set the display color of the corresponding curve. Currently, all eight buttons are in the "show" state, so the above eight test curves will be displayed in the vacuum test graph in the middle of the screen. The "show/hide" status of each switch button is set appropriately, allowing the user to analyze the relationship between various curve changes.

To analyze the relationship between the change of micro-oxygen concentration and the cycling cleaning regeneration phase, users can display five test curves such as TT-A, TT-B, OT-2, regeneration temperature setting, and OT-2 setting in this screen, as shown in **Figure 16(b)**. It can be seen from the figure that the reaction tank A performs cleaning for the T6 time (OT-2 curve drop) and rests for T7 time (OT-2 curve is flat). After two such cycles, the reaction tank B is immediately switched to perform such a cleaning rest cycle (the change in the OT-2 curve from continuing to fall till becoming flat), while the reaction tank A simultaneously

**53**

**6. Conclusions**

**Figure 16.**

*The Design and Development of Control System for High Vacuum Deoxygenated...*

starts the regeneration operation, that is, it is heated to the regeneration temperature (the orange TT-A curve rises to the regeneration temperature setting), and then 3% hydrogen is entered to regenerate the cleaning reagent. After the reaction tank B performs two cycles of cleaning rest, it will switch back to the reaction tank A to continue the cycling cleaning operation, when the reaction tank B will start the regenerating operation synchronously, because the red TT-B curve will rise due to heating the regeneration temperature setting. From the above changes in the curves, it can be seen that the operation of removing the micro-oxygen from the glove box can be performed in turn by two reaction tanks, and no interruption occurs due to the run out of the reagent of a reaction tank. In addition, for the change of the micro-oxygen concentration, performing two cleaning cycles in the tank A can reduce the micro-oxygen concentration from 1000 ppm to about 390 ppm. And two cycling cleaning operations by the reaction tank B is followed, and the micro-

*Display screen of cycling cleaning test curves. (a) Complete test curve of the cycling cleaning regeneration* 

This article develops an advanced control system for high vacuum deoxygenation and water removal for the glove box. This system can use the switching of the hardware pipeline and the control of monitoring software to provide users with the transfer box cleaning, box body cleaning, vacuum preservation, end preservation, hydrogen storage, hydrogen cell cleaning and cycling cleaning regeneration operating functions, and each operating function can be automatically executed through the settings of control parameters and instrument parameters. This system combined sequential control with multi-conditional logic control. In the state transition diagram of control process, there are mainly the timing control (according to the timing parameters of T1, T2, … or T9) and multiple characteristic monitoring controls (such as temperature, vacuum pressure, oxygen concentration, water concentration, … and other control parameters). The system provides users to set the target values of parameters for the sake of multi-conditional monitoring and logic control. Hence, the system may decide the direction of execution when facing with various events. As a result, the control output of multi-conditional judgment may be highly tolerant and stable toward signal offset and change in other external

oxygen concentration can be further reduced to about 210 ppm.

conditions. Therefore, this control system may be high robustness.

From the control point of view, the cycling cleaning regeneration function provided by the system must be considered in terms of its control, including the comparison between the detection values of water and oxygen concentration and

*DOI: http://dx.doi.org/10.5772/intechopen.80423*

*operation and (b) relationship among TT-A, TT-B, and OT-2.*

*The Design and Development of Control System for High Vacuum Deoxygenated... DOI: http://dx.doi.org/10.5772/intechopen.80423*

**Figure 16.**

*Applied Modern Control*

**Figure 15.**

*relationship between PT-1 and HT-1.*

the box, you may click "TT-1" and "OT-1" switch buttons to set them to the "hide" status, and make the other three buttons "HT-1," "PT-1," and "PT-1 setup" maintain the "show" state, as shown in **Figure 15(b)**, which is the display of the relationship between the PT-1 and HT-1 curves. From this screen, it is observed that the green PT-1 pressure curve causes the significant drop of the blue HT-1 water vapor concentration during each descent (which represents the vacuumizing stage), and whether the PT-1 pressure curve can reach the gray PT-1 setup curve (i.e., reaching the setting value of the vacuum pressure) at each descent; and the pressure curve of PT-1 during the flat bottom is the nitrogen filling phase of the box body, when the change in the HT-1 water vapor concentration curve is less obvious. From the above analysis, we can see that the effect of the vacuumizing stage on the decline of water vapor concentration is obviously higher than that of the nitrogen filling stage.

*Display screen of vacuum test curves of box body cleaning. (a) Complete vacuum test curve and (b) change* 

By using automatically recorded box temperature TT-1, regeneration temperature TT-A of the reaction tank A, regeneration temperature TT-B of the reaction tank B, the micro-water concentration HT-2 and micro-oxygen concentration OT-2 detected by the sampling tank, and the settings of HT-2 and OT-2, the system can draw the clean test curves of the cycling cleaning regeneration operation, as shown in **Figure 16**, which is the elastic display screen of the test curve of such cleaning. **Figure 16(a)** shows the complete cleaning test curves. There are eight switching buttons left side of the screen, such as "TT-1," "TT-A," "TT-B," "HT-2," "OT-2," "Regen Temp setup," "OT-2 setup," and "HT-2 setup" and eight corresponding curve color setting blocks. Each switching button can be used by the user to switch between show/hide, and each color setting box enables the user to set the display color of the corresponding curve. Currently, all eight buttons are in the "show" state, so the above eight test curves will be displayed in the vacuum test graph in the middle of the screen. The "show/hide" status of each switch button is set appropriately, allowing the user to analyze the relationship between various curve changes. To analyze the relationship between the change of micro-oxygen concentration and the cycling cleaning regeneration phase, users can display five test curves such as TT-A, TT-B, OT-2, regeneration temperature setting, and OT-2 setting in this screen, as shown in **Figure 16(b)**. It can be seen from the figure that the reaction tank A performs cleaning for the T6 time (OT-2 curve drop) and rests for T7 time (OT-2 curve is flat). After two such cycles, the reaction tank B is immediately switched to perform such a cleaning rest cycle (the change in the OT-2 curve from continuing to fall till becoming flat), while the reaction tank A simultaneously

**5.2 Cleaning test curve rendering analysis**

**52**

*Display screen of cycling cleaning test curves. (a) Complete test curve of the cycling cleaning regeneration operation and (b) relationship among TT-A, TT-B, and OT-2.*

starts the regeneration operation, that is, it is heated to the regeneration temperature (the orange TT-A curve rises to the regeneration temperature setting), and then 3% hydrogen is entered to regenerate the cleaning reagent. After the reaction tank B performs two cycles of cleaning rest, it will switch back to the reaction tank A to continue the cycling cleaning operation, when the reaction tank B will start the regenerating operation synchronously, because the red TT-B curve will rise due to heating the regeneration temperature setting. From the above changes in the curves, it can be seen that the operation of removing the micro-oxygen from the glove box can be performed in turn by two reaction tanks, and no interruption occurs due to the run out of the reagent of a reaction tank. In addition, for the change of the micro-oxygen concentration, performing two cleaning cycles in the tank A can reduce the micro-oxygen concentration from 1000 ppm to about 390 ppm. And two cycling cleaning operations by the reaction tank B is followed, and the microoxygen concentration can be further reduced to about 210 ppm.

#### **6. Conclusions**

This article develops an advanced control system for high vacuum deoxygenation and water removal for the glove box. This system can use the switching of the hardware pipeline and the control of monitoring software to provide users with the transfer box cleaning, box body cleaning, vacuum preservation, end preservation, hydrogen storage, hydrogen cell cleaning and cycling cleaning regeneration operating functions, and each operating function can be automatically executed through the settings of control parameters and instrument parameters. This system combined sequential control with multi-conditional logic control. In the state transition diagram of control process, there are mainly the timing control (according to the timing parameters of T1, T2, … or T9) and multiple characteristic monitoring controls (such as temperature, vacuum pressure, oxygen concentration, water concentration, … and other control parameters). The system provides users to set the target values of parameters for the sake of multi-conditional monitoring and logic control. Hence, the system may decide the direction of execution when facing with various events. As a result, the control output of multi-conditional judgment may be highly tolerant and stable toward signal offset and change in other external conditions. Therefore, this control system may be high robustness.

From the control point of view, the cycling cleaning regeneration function provided by the system must be considered in terms of its control, including the comparison between the detection values of water and oxygen concentration and the set values, whether the cleaning capacity of the reaction tank during operation is already insufficient, whether the regeneration reaction tank has completed the regeneration time, and so on. The monitoring host can use these judgment results to simultaneously perform the cycling cleaning and cycling regeneration control. This kind of control mode is a double loop control mode, in which cleaning and regeneration are performed at the same time. On the other hand, during the execution of various operating functions, the system can automatically record all measured data and display the vacuum test curves and cleaning test curves flexibly. This can provide the user with the analysis of performance on the glove box cleaning operation and vacuum operation, and the analysis results can be used as a basis for determining the quality parameters of the glove box and can also be used as a basis for the glove box user to plan experiments or operating procedures.

### **Acknowledgements**

This study was sponsored by the project of Ministry of Science and Technology, R.O.C., project number: MOST 106-2221-E-344-004.

### **Author details**

Ming-Sen Hu Department of Aviation and Communication Electronics, Air Force Institute of Technology, Kaohsiung City, Taiwan, R.O.C.

\*Address all correspondence to: mshu1227@gmail.com

© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

**55**

*DOI: http://dx.doi.org/10.5772/intechopen.80423*

[1] Ashby EC, Schwartz RD. A glove BOX system for the manipulation of air sensitive compounds. Journal of Chemical Education. 1974;**51**(1):65-68

[2] Boylen CW. Installation and use of the microscope within a gastight glove Box. Journal of Applied Microbiology.

[3] Cournoyer ME, Castro JM, Lee MB,

Lawton CM, Ho Park Y, Lee R, Schreiber S. Elements of a glovebox glove integrity program. Journal of Chemical Health and Safety. 2009;

[4] Dryburgh PM. A glove-box and drying system for the manipulation of moisture sensitive materials. Journal of Scientific Instruments.

[5] Smith C, Mullins D, Nakaoka R. A novel process control system for a glove box vitrification system. In: WM'04 Conference. 2004

[6] Su CS. Vacuum Technology Essence. Taipei, Taiwan, R.O.C: Wu Nan Book Corp.; 2013. ISBN13: 9789571134789

[7] Wu SC, Wang RW, Lin MY. Vacuum Technology and Applications. Taipei, Taiwan, R.O.C: San Ming Book Corp.;

[8] Cheng HB, Chu CJ. Painting Vacuum Technology. Taipei, Taiwan, R.O.C:

[9] Campbell J. Speed Cleaning. New

[10] National Laboratory Research Institute of Instrument Science and Technology. Vacuum Technology and Applications. New Taipei City, Taiwan, R.O.C: Chuan Yua Book Corp.; 2004.

2001. ISBN13: 9789570286762

Shu Chuan Publisher; 2012

York: Rodale; 2005. p. 97. ISBN 1-59486-274-5

ISBN: 9570286768

1971;**21**(6):1089-1090

**References**

1967;**44**(8):658-660

**16**(1):4-10

*The Design and Development of Control System for High Vacuum Deoxygenated...*

[11] Cournoyer ME, Lee MB, Schreiber S. Minimizing glovebox glove failures, Part IV: Control charts, LA-UR 07-0783. Journal of the American Society of Mechanical Engineers. In: Proceeding from WM'07; 25 February-1 Mar;

[12] Smith JD. The Virtual Glovebox (VGX): An immersive simulation system for training astronauts to perform glovebox experiments in space.

Tucson, Arizona. 2007

Enclosure. 2002;**15**:8-13

[13] Hu MS. A high vacuum deoxygenation and water removal glove box control system. In: Aviation

Technology and Fight Safety

San Antonio, Texas. 2006

February; Tucson, AZ. 2002

[16] Glove Box Workstations. MBRAUN. Available from:

glovebox-workstations

http://www.mbraun.com/products/

[17] Michael EC, Stephen S. Glove Testing & Evaluation, LAUR 07-3150,

JOWOG 30 Exchange, Atomic Weapons Establishment; 21-25 May;

Aldermaston, England. 2007

[18] Hu MS, Su RH. A glove box measuring control system and its measuring control method. Patent of the Republic of China, patent number: I594854, duration: 2/2017 to 1/2037

[19] Hu MS, Chen LH. The development of a high vacuum glove box control

Conference; June 9; Haoshuing. 2017

[14] Glovebox Glove Improvement Workshop, LA-UR 06-5176. In: American Glovebox Society Annual Conference 2005; July 16-20;

[15] Uematsu S, Kashiro K, Tobita N. Dismantling of gloveboxes for MOX fuel fabrication by a glovebox dismantling facility. In: WM'02 Conference; 24-28

*DOI: http://dx.doi.org/10.5772/intechopen.80423 The Design and Development of Control System for High Vacuum Deoxygenated...*

#### **References**

*Applied Modern Control*

**Acknowledgements**

**54**

**Author details**

Ming-Sen Hu

provided the original work is properly cited.

Technology, Kaohsiung City, Taiwan, R.O.C.

\*Address all correspondence to: mshu1227@gmail.com

© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium,

Department of Aviation and Communication Electronics, Air Force Institute of

the set values, whether the cleaning capacity of the reaction tank during operation is already insufficient, whether the regeneration reaction tank has completed the regeneration time, and so on. The monitoring host can use these judgment results to simultaneously perform the cycling cleaning and cycling regeneration control. This kind of control mode is a double loop control mode, in which cleaning and regeneration are performed at the same time. On the other hand, during the execution of various operating functions, the system can automatically record all measured data and display the vacuum test curves and cleaning test curves flexibly. This can provide the user with the analysis of performance on the glove box cleaning operation and vacuum operation, and the analysis results can be used as a basis for determining the quality parameters of the glove box and can also be used as a basis

This study was sponsored by the project of Ministry of Science and Technology,

for the glove box user to plan experiments or operating procedures.

R.O.C., project number: MOST 106-2221-E-344-004.

[1] Ashby EC, Schwartz RD. A glove BOX system for the manipulation of air sensitive compounds. Journal of Chemical Education. 1974;**51**(1):65-68

[2] Boylen CW. Installation and use of the microscope within a gastight glove Box. Journal of Applied Microbiology. 1971;**21**(6):1089-1090

[3] Cournoyer ME, Castro JM, Lee MB, Lawton CM, Ho Park Y, Lee R, Schreiber S. Elements of a glovebox glove integrity program. Journal of Chemical Health and Safety. 2009; **16**(1):4-10

[4] Dryburgh PM. A glove-box and drying system for the manipulation of moisture sensitive materials. Journal of Scientific Instruments. 1967;**44**(8):658-660

[5] Smith C, Mullins D, Nakaoka R. A novel process control system for a glove box vitrification system. In: WM'04 Conference. 2004

[6] Su CS. Vacuum Technology Essence. Taipei, Taiwan, R.O.C: Wu Nan Book Corp.; 2013. ISBN13: 9789571134789

[7] Wu SC, Wang RW, Lin MY. Vacuum Technology and Applications. Taipei, Taiwan, R.O.C: San Ming Book Corp.; 2001. ISBN13: 9789570286762

[8] Cheng HB, Chu CJ. Painting Vacuum Technology. Taipei, Taiwan, R.O.C: Shu Chuan Publisher; 2012

[9] Campbell J. Speed Cleaning. New York: Rodale; 2005. p. 97. ISBN 1-59486-274-5

[10] National Laboratory Research Institute of Instrument Science and Technology. Vacuum Technology and Applications. New Taipei City, Taiwan, R.O.C: Chuan Yua Book Corp.; 2004. ISBN: 9570286768

[11] Cournoyer ME, Lee MB, Schreiber S. Minimizing glovebox glove failures, Part IV: Control charts, LA-UR 07-0783. Journal of the American Society of Mechanical Engineers. In: Proceeding from WM'07; 25 February-1 Mar; Tucson, Arizona. 2007

[12] Smith JD. The Virtual Glovebox (VGX): An immersive simulation system for training astronauts to perform glovebox experiments in space. Enclosure. 2002;**15**:8-13

[13] Hu MS. A high vacuum deoxygenation and water removal glove box control system. In: Aviation Technology and Fight Safety Conference; June 9; Haoshuing. 2017

[14] Glovebox Glove Improvement Workshop, LA-UR 06-5176. In: American Glovebox Society Annual Conference 2005; July 16-20; San Antonio, Texas. 2006

[15] Uematsu S, Kashiro K, Tobita N. Dismantling of gloveboxes for MOX fuel fabrication by a glovebox dismantling facility. In: WM'02 Conference; 24-28 February; Tucson, AZ. 2002

[16] Glove Box Workstations. MBRAUN. Available from: http://www.mbraun.com/products/ glovebox-workstations

[17] Michael EC, Stephen S. Glove Testing & Evaluation, LAUR 07-3150, JOWOG 30 Exchange, Atomic Weapons Establishment; 21-25 May; Aldermaston, England. 2007

[18] Hu MS, Su RH. A glove box measuring control system and its measuring control method. Patent of the Republic of China, patent number: I594854, duration: 2/2017 to 1/2037

[19] Hu MS, Chen LH. The development of a high vacuum glove box control

system with cycling cleaning and regeneration. In: IEEE International Conference on Systems Science and Engineering; 21-23 July; Vietnam. 2017

[20] Advantech USB-4711A User Manual, 150 KS/s, 12-bit Multifunction USB Data Acquisition Module, Advantech

[21] USB-4750 User Manual, 32-Channel Isolated Digtial I/O USB Module, Advantech

[22] Hsiao TC, Wang CY, Chu JW. Virtual Instrument Control Programming—LabVIEW 8X. New Taipei City, Taiwan, R.O.C: Gao Li Book Corp.; 2014

[23] Chen CC. LabVIEW Applications (Include Automatic Measurement and Remote Control). New Taipei City, Taiwan, R.O.C: Chuan Yua Book Corp.; 2014

[24] Hsu YH. Interface Design and Practice-Using LabVIEW. New Taipei City, Taiwan, R.O.C: Chuan Yua Book Corp; 2014

[25] Goodwin GC, Graebe SF, Salgado ME. Control System Design. New Jersey: Prentice Hall; 2001

[26] Kuo BC. Automatic Control Systems. New York: Holt, Rinehart and Winston, Inc.; 2003

**57**

**Chapter 4**

**Abstract**

**1. Introduction**

by the conventional controller.

Column

*Rajeev Kumar Dohare*

Artificial Intelligent-Based

as feed flow rate, feed compositions, and liquid split factor.

columns. The schematic diagram of the DWC is shown in **Figure 1**.

**Keywords:** artificial intelligent, DWC, BTX, distillation, simulation, MATLAB

Divided wall column is the combination of four thermally coupled distillation

Benzene-toluene-o-xylene (BTX) system as feed is the dividing wall column for separation purpose. Feed introduced into the prefractionator side of the wall. A side stream is removed from the other side. The side stream is mostly the intermediate boiling component of the ternary mixture. The lightest component (benzene) goes overhead in the distillate product, and the heaviest component (o-xylene) goes out in the bottom product. Therefore, a single dividing wall column can separate a ternary mixture into three pure product streams. Due to high interactions among the process variables and due to restricted experiences, it is very difficult to control

In many chemical processes, artificial neural network has been implemented successfully in the process identification and control of the nonlinear dynamic systems. This is the main reason to switch over to the recent control strategies such as artificial neural network, fuzzy logic, etc. All such control technique-based

Predictive Control of Divided Wall

Distillation is the most popular thermal separation technique used in the chemical and petrochemical process industry for the liquid mixture separation. Certainly, the distillation has a plenty of advantages, yet it has a drawback such as more energy requirement. In order to reduce the energy consumption of the conventional distillation column, energy integration is applied within the distillation column. The most important thermally coupled distillation column sequence is the Petlyuk column, which uses two recycle streams. Petlyuk column is a novel design that integrates two distillation columns into one shell, which is known as a dividing wall column. A dividing wall column (DWC) offers the possibility to separate a multicomponent mixture into high-purity products or sharp splits. Artificial neural network predictive controller (ANNPC) has been implemented to control the DWC. The performance of the ANNPC control strategies and the dynamic response of the DWC are investigated in terms of the product composition in the different sections of the dividing wall column for the different persistent disturbances such

[27] Nise NS. Control System Engineering. 4th ed. New York: John Wiley & Sons, Inc.; 2004

#### **Chapter 4**

*Applied Modern Control*

system with cycling cleaning and regeneration. In: IEEE International Conference on Systems Science and Engineering; 21-23 July; Vietnam. 2017

[20] Advantech USB-4711A User Manual, 150 KS/s, 12-bit Multifunction

USB Data Acquisition Module,

[22] Hsiao TC, Wang CY, Chu JW. Virtual Instrument

Book Corp.; 2014

2014

Corp; 2014

Prentice Hall; 2001

Winston, Inc.; 2003

[21] USB-4750 User Manual, 32-Channel Isolated Digtial I/O USB Module,

Control Programming—LabVIEW 8X. New Taipei City, Taiwan, R.O.C: Gao Li

[23] Chen CC. LabVIEW Applications (Include Automatic Measurement and Remote Control). New Taipei City, Taiwan, R.O.C: Chuan Yua Book Corp.;

[24] Hsu YH. Interface Design and Practice-Using LabVIEW. New Taipei City, Taiwan, R.O.C: Chuan Yua Book

[25] Goodwin GC, Graebe SF, Salgado ME. Control System Design. New Jersey:

[26] Kuo BC. Automatic Control Systems. New York: Holt, Rinehart and

[27] Nise NS. Control System Engineering. 4th ed. New York: John Wiley & Sons, Inc.; 2004

Advantech

Advantech

**56**

## Artificial Intelligent-Based Predictive Control of Divided Wall Column

*Rajeev Kumar Dohare*

#### **Abstract**

Distillation is the most popular thermal separation technique used in the chemical and petrochemical process industry for the liquid mixture separation. Certainly, the distillation has a plenty of advantages, yet it has a drawback such as more energy requirement. In order to reduce the energy consumption of the conventional distillation column, energy integration is applied within the distillation column. The most important thermally coupled distillation column sequence is the Petlyuk column, which uses two recycle streams. Petlyuk column is a novel design that integrates two distillation columns into one shell, which is known as a dividing wall column. A dividing wall column (DWC) offers the possibility to separate a multicomponent mixture into high-purity products or sharp splits. Artificial neural network predictive controller (ANNPC) has been implemented to control the DWC. The performance of the ANNPC control strategies and the dynamic response of the DWC are investigated in terms of the product composition in the different sections of the dividing wall column for the different persistent disturbances such as feed flow rate, feed compositions, and liquid split factor.

**Keywords:** artificial intelligent, DWC, BTX, distillation, simulation, MATLAB

#### **1. Introduction**

Divided wall column is the combination of four thermally coupled distillation columns. The schematic diagram of the DWC is shown in **Figure 1**.

Benzene-toluene-o-xylene (BTX) system as feed is the dividing wall column for separation purpose. Feed introduced into the prefractionator side of the wall. A side stream is removed from the other side. The side stream is mostly the intermediate boiling component of the ternary mixture. The lightest component (benzene) goes overhead in the distillate product, and the heaviest component (o-xylene) goes out in the bottom product. Therefore, a single dividing wall column can separate a ternary mixture into three pure product streams. Due to high interactions among the process variables and due to restricted experiences, it is very difficult to control by the conventional controller.

In many chemical processes, artificial neural network has been implemented successfully in the process identification and control of the nonlinear dynamic systems. This is the main reason to switch over to the recent control strategies such as artificial neural network, fuzzy logic, etc. All such control technique-based

**Figure 1.** *Schematic diagram of divided wall distillation column.*

controllers are known as intelligent controllers. Settling time in the model-based controller is less in comparison to the conventional controller without compromising the product purities. To further improve the control performance, the artificial intelligence techniques were also attempted.

Artificial neural networks (ANN) have been designed on the complexities of the brain functions in an effort to capture the amazing learning capabilities of the brain as shown in the **Figure 2**. ANN is a parallel computer or processor designed to imitate the way the brain accomplishes a certain task [1]. The smallest processing element of ANN is a neuron or node, which helps to do sample calculations. Using the neurons collectively with massive connections among them results in a network that is able to process and store relative information for mapping the network inputs to its outputs. With this feasibility and its capability, there are most widespread

**59**

*Artificial Intelligent-Based Predictive Control of Divided Wall Column*

interests in solving complicate problems particularly in the fields of pattern recognition, control, forecasting, classification, system identification, and optimization. Authors already discussed, about the desired product purity at the optimized operating parameters viz. feed tray location, reflux ratio, liquid reflux rate, vapor split ratio, etc. [2]. To control the dividing wall column is the big challenge for the control engineers due to its high dynamics. Conventional controller like PID is not performed well in comparison to the advanced controller which was shown by the authors in their research [3]. Artificial neural network has been applied on dividing wall column for controlling the tray temperatures that is based on predictive technique. Composition of the product in the dividing wall column is the main controlled variable but due to the measurement delay in the online analyzer, it is rarely used. For successful operation, monitoring, and controlling chemical process, an accurate online analyzer of important quality variables is essential. However, the measurement of all these variables online is a big task due to the limitations such as the high cost, time delay, and reliability, therefore they cannot be directly close-loop controlled. Composition is controlled indirectly by controlling the temperature of the different sections in DWC. To enhance the controllability, artificial neural network-based predictive controller (ANNPC) can be used to control the temperature of the different sections

In actual industrial circumstances, it would be ensured that the equipment should be run safely and efficiently with the important process variables that relate mostly with system stabilizations and product qualities have to be controlled in real time. However, it is very difficult to measure these variables by online physical sensors and its economic issue [4]. Due to the importance of the control problem, many methods have been adopted in the past: the first method is a quality open-loop control, in which to get the quality of the products, the undue purification is required; the second method is useful indirectly to control the quality of close-loop, such as controlling the temperature at each plate in the column at different sections. But sometimes, this method cannot control the quality of the product; third method is online process gas chromatography. Due to its limitations like cost, reliability, and

the time delay, it cannot satisfy the online control requirement of quality.

The artificial neural network predictive control was trained using Levenberg-Marquardt optimization algorithm. Hidden neurons use the sigmoid activation function, whereas output layer neurons use the linear activation function. The selection of the different layer neurons depends on the complexity of the problem. On a trial-and-error basis, a number of neurons were selected in this present study. The general structure of the neural network is shown in **Figure 1**. For the normalization, all the inputs and outputs should be dealt up with different magnitudes. The total number of data is divided into different parts for neural network model building: (1) training data, (2) validation data, and (3) testing data. First part of the data is used for training purpose, the second part for validation of the network structure, and third part of the data is useful to evaluate the selected model.

Artificial neural network predictive control (ANNPC) is a combination of artificial neural network- and model-based controller. The multilayer

*DOI: http://dx.doi.org/10.5772/intechopen.81261*

in the dividing wall column.

**2. MIMO neural network generation**

**3. Artificial neural network predictive controller**

#### *Artificial Intelligent-Based Predictive Control of Divided Wall Column DOI: http://dx.doi.org/10.5772/intechopen.81261*

*Applied Modern Control*

**58**

**Figure 1.**

**Figure 2.** *ANN structure.*

controllers are known as intelligent controllers. Settling time in the model-based controller is less in comparison to the conventional controller without compromising the product purities. To further improve the control performance, the artificial

Artificial neural networks (ANN) have been designed on the complexities of the brain functions in an effort to capture the amazing learning capabilities of the brain as shown in the **Figure 2**. ANN is a parallel computer or processor designed to imitate the way the brain accomplishes a certain task [1]. The smallest processing element of ANN is a neuron or node, which helps to do sample calculations. Using the neurons collectively with massive connections among them results in a network that is able to process and store relative information for mapping the network inputs to its outputs. With this feasibility and its capability, there are most widespread

intelligence techniques were also attempted.

*Schematic diagram of divided wall distillation column.*

interests in solving complicate problems particularly in the fields of pattern recognition, control, forecasting, classification, system identification, and optimization.

Authors already discussed, about the desired product purity at the optimized operating parameters viz. feed tray location, reflux ratio, liquid reflux rate, vapor split ratio, etc. [2]. To control the dividing wall column is the big challenge for the control engineers due to its high dynamics. Conventional controller like PID is not performed well in comparison to the advanced controller which was shown by the authors in their research [3]. Artificial neural network has been applied on dividing wall column for controlling the tray temperatures that is based on predictive technique. Composition of the product in the dividing wall column is the main controlled variable but due to the measurement delay in the online analyzer, it is rarely used. For successful operation, monitoring, and controlling chemical process, an accurate online analyzer of important quality variables is essential. However, the measurement of all these variables online is a big task due to the limitations such as the high cost, time delay, and reliability, therefore they cannot be directly close-loop controlled. Composition is controlled indirectly by controlling the temperature of the different sections in DWC. To enhance the controllability, artificial neural network-based predictive controller (ANNPC) can be used to control the temperature of the different sections in the dividing wall column.

In actual industrial circumstances, it would be ensured that the equipment should be run safely and efficiently with the important process variables that relate mostly with system stabilizations and product qualities have to be controlled in real time. However, it is very difficult to measure these variables by online physical sensors and its economic issue [4]. Due to the importance of the control problem, many methods have been adopted in the past: the first method is a quality open-loop control, in which to get the quality of the products, the undue purification is required; the second method is useful indirectly to control the quality of close-loop, such as controlling the temperature at each plate in the column at different sections. But sometimes, this method cannot control the quality of the product; third method is online process gas chromatography. Due to its limitations like cost, reliability, and the time delay, it cannot satisfy the online control requirement of quality.

#### **2. MIMO neural network generation**

The artificial neural network predictive control was trained using Levenberg-Marquardt optimization algorithm. Hidden neurons use the sigmoid activation function, whereas output layer neurons use the linear activation function. The selection of the different layer neurons depends on the complexity of the problem. On a trial-and-error basis, a number of neurons were selected in this present study. The general structure of the neural network is shown in **Figure 1**. For the normalization, all the inputs and outputs should be dealt up with different magnitudes. The total number of data is divided into different parts for neural network model building: (1) training data, (2) validation data, and (3) testing data. First part of the data is used for training purpose, the second part for validation of the network structure, and third part of the data is useful to evaluate the selected model.

#### **3. Artificial neural network predictive controller**

Artificial neural network predictive control (ANNPC) is a combination of artificial neural network- and model-based controller. The multilayer

**Figure 3.**

*Line diagram of neural network predictive control.*

perceptron makes it a good choice for modeling nonlinear systems and for implementing nonlinear controller. The line diagram for the use of a neural network implemented on a process is shown in **Figure 3**. The unknown function may correspond to a controlled system, and the neural network is the identified plant model. Two-layer networks, with sigmoid transfer functions in the hidden layer and linear transfer functions in the output layer, are universal approximations. Artificial neural network is considered here as a predictive model for controlling the BTX system. The ultimate achievement of the model-based predictive controller is to generate a sequence of control signals minimizing a cost index that is the function of difference between the future process outputs from the desired set points and control moves. The ANN control structure was designed in "nntool" box, which was exported in Simulink environment to control DWC. An S-function was written in MATLAB for representing the DWC model. This DWC model was integrated with neural network control structure in MATLAB.

#### **4. Data generation for training, testing, and validation of the network**

To generate training data for the neural network training, a uniform random number generator for all the three input variables has been taken in the Simulink model as shown in **Figure 4**. The random number was kept constant at least for 1000 s. About 700 samples of each variable were generated with a sampling period of 100 s. These entire sample data were divided into three segments, viz.: training data, validation data, and testing data, in the ratio of 70:15:15, respectively.

The output performance of neural network depends on the number of neurons in input layer, number of hidden layer and optimizing algorithms. For the better output performance of the neural network, 90 neurons have been chosen in the input layer. The 90-input neurons correspond to 15 past values of each input and output variables. The numbers of neurons in hidden layer were found by the performance index such as R2. The performance analysis of the neural network is shown in **Table 1**.

**61**

**Figure 4.**

*subsystem of subsystem1 "SuccDelay."*

*Systematic Simulink model of the dynamic DWC for data generation. (a) Main figure, (b) subsystem1, and (c)* 

*Artificial Intelligent-Based Predictive Control of Divided Wall Column*

*DOI: http://dx.doi.org/10.5772/intechopen.81261*

*Artificial Intelligent-Based Predictive Control of Divided Wall Column DOI: http://dx.doi.org/10.5772/intechopen.81261*

*Applied Modern Control*

perceptron makes it a good choice for modeling nonlinear systems and for implementing nonlinear controller. The line diagram for the use of a neural network implemented on a process is shown in **Figure 3**. The unknown function may correspond to a controlled system, and the neural network is the identified plant model. Two-layer networks, with sigmoid transfer functions in the hidden layer and linear transfer functions in the output layer, are universal approximations. Artificial neural network is considered here as a predictive model for controlling the BTX system. The ultimate achievement of the model-based predictive controller is to generate a sequence of control signals minimizing a cost index that is the function of difference between the future process outputs from the desired set points and control moves. The ANN control structure was designed in "nntool" box, which was exported in Simulink environment to control DWC. An S-function was written in MATLAB for representing the DWC model. This DWC model was integrated with neural network control structure

**4. Data generation for training, testing, and validation of the network**

To generate training data for the neural network training, a uniform random number generator for all the three input variables has been taken in the Simulink model as shown in **Figure 4**. The random number was kept constant at least for 1000 s. About 700 samples of each variable were generated with a sampling period of 100 s. These entire sample data were divided into three segments, viz.: training data, validation data, and testing data, in the ratio of 70:15:15,

The output performance of neural network depends on the number of neurons

in input layer, number of hidden layer and optimizing algorithms. For the better output performance of the neural network, 90 neurons have been chosen in the input layer. The 90-input neurons correspond to 15 past values of each input and output variables. The numbers of neurons in hidden layer were found by the performance index such as R2. The performance analysis of the neural network is

**60**

in MATLAB.

**Figure 3.**

*Line diagram of neural network predictive control.*

respectively.

shown in **Table 1**.

#### **Figure 4.**

*Systematic Simulink model of the dynamic DWC for data generation. (a) Main figure, (b) subsystem1, and (c) subsystem of subsystem1 "SuccDelay."*


**Table 1.**

*Performance parameters of neural network.*

#### **5. Neural network training and its algorithm**

Delgado et al. [5] suggested controlling the process online by the neural network; it requires accurate network training data to design some network frames, so that the model has better extrapolation and suaveness ability [5].

**63**

**Figure 6.**

*Artificial Intelligent-Based Predictive Control of Divided Wall Column*

When artificial neural network technique is applied on any process model, the common method is to collect the data for training either directly from the process plant or from the simulation. The training datasets for the network generation in the neural network have been obtained by the open-loop process under dynamic operating conditions because all the steady-state variables are in nonoscillatory motion. Due to nonavailability of the experimental BTX data of dividing wall distillation column, a mathematical model has been developed and then simulated in real time to find out the training data for network generation. Backpropagation algorithm has been used as a training algorithm to tune the connection weights for the function of each neuron. For the offline training of the neural network, different types of training algorithms such as Levenberg-Marquardt, gradient-descent, and conjugate gradient can be used. But Levenberg-Marquardt algorithm is a fast-converging optimization technique among all. For DWC, the training, testing, and validation results are shown in **Figure 5**.

The results show that there is a good fit between actual data and predicted data.

[6]. It is an easy way to improve the calculation up to the desired precision.

**6. Simulink model of artificial neural network predictive controller**

A Simulink model was designed to analyze the control performance of the artificial neural network predictive controller on the various load disturbance parameters. A systematic block diagram of the artificial neural network predictive

This algorithm works on the gradient and Hessian matrix of the objective function. In the Levenberg-Marquardt algorithm, Hessian matrix is approximated by Jacobian matrix and it is calculated by second-order derivative of objective function

*DOI: http://dx.doi.org/10.5772/intechopen.81261*

controller is given in **Figure 6**.

**7. Results and discussions**

**7.1 Effect of benzene composition change in feed**

*Systematic block diagram of artificial neural network predictive control.*

Load change of ±10% in benzene composition of the feed was given to check the performance of ANNPC as shown in **Figure 7**. The controller was able to bring

**Figure 5.** *Training, testing, and validation of the generated data of DWC.*

*Artificial Intelligent-Based Predictive Control of Divided Wall Column DOI: http://dx.doi.org/10.5772/intechopen.81261*

When artificial neural network technique is applied on any process model, the common method is to collect the data for training either directly from the process plant or from the simulation. The training datasets for the network generation in the neural network have been obtained by the open-loop process under dynamic operating conditions because all the steady-state variables are in nonoscillatory motion. Due to nonavailability of the experimental BTX data of dividing wall distillation column, a mathematical model has been developed and then simulated in real time to find out the training data for network generation. Backpropagation algorithm has been used as a training algorithm to tune the connection weights for the function of each neuron.

For the offline training of the neural network, different types of training algorithms such as Levenberg-Marquardt, gradient-descent, and conjugate gradient can be used. But Levenberg-Marquardt algorithm is a fast-converging optimization technique among all. For DWC, the training, testing, and validation results are shown in **Figure 5**. The results show that there is a good fit between actual data and predicted data.

This algorithm works on the gradient and Hessian matrix of the objective function. In the Levenberg-Marquardt algorithm, Hessian matrix is approximated by Jacobian matrix and it is calculated by second-order derivative of objective function [6]. It is an easy way to improve the calculation up to the desired precision.

#### **6. Simulink model of artificial neural network predictive controller**

A Simulink model was designed to analyze the control performance of the artificial neural network predictive controller on the various load disturbance parameters. A systematic block diagram of the artificial neural network predictive controller is given in **Figure 6**.

**Figure 6.**

*Applied Modern Control*

**Performance function**

Mean squared normalized error

*Performance parameters of neural network.*

**Neurons in hidden layer**

**Training algorithm**

Levenberg-Marquardt backpropagation

**Table 1.**

**5. Neural network training and its algorithm**

that the model has better extrapolation and suaveness ability [5].

Delgado et al. [5] suggested controlling the process online by the neural network; it requires accurate network training data to design some network frames, so

**Epoch Max** 

**validation check**

 200 100 1e-7 0.93 0.95 0.96 150 100 1e-7 0.96 0.88 0.93 100 100 1e-7 0.94 0.92 0.95 100 100 1e-6 0.95 0.91 0.94 50 25 1e-6 0.95 0.94 0.93 50 25 1e-7 0.93 0.97 0.98 20 10 1e-7 0.94 0.96 0.97

**Tolerance Training Validation Testing**

**62**

**Figure 5.**

*Training, testing, and validation of the generated data of DWC.*

*Systematic block diagram of artificial neural network predictive control.*

#### **7. Results and discussions**

#### **7.1 Effect of benzene composition change in feed**

Load change of ±10% in benzene composition of the feed was given to check the performance of ANNPC as shown in **Figure 7**. The controller was able to bring the temperatures in all the sections to the desired set points without any offset. The settling time in section 1 and section 4 for ±10% change in benzene composition is nearly 0.6 and 0.45 h, respectively. Although main column has 0.32 and 0.42 h settling time for −10% change and +10% change in benzene composition, respectively. The maximum peak of temperature in the section one (Tsec1) is 0.4 and 0.28°C at −10% change and +10% change, respectively. Section 3 temperature (Tsec3) and section 4 temperature (Tsec4) have very small peak value in the range of 10<sup>−</sup><sup>3</sup> and 10<sup>−</sup><sup>1</sup> °C for the ±10% change. With respect to the temperature change of the section 1, benzene composition reaches to the desired set point without any offset in 0.6 and 0.7 h for −10% and +10% change. Moreover, toluene and xylene composition acquires the desired value in 0.5 and 0.4 h, respectively without showing moderate offset and small peaks (i.e., in the range of 10<sup>−</sup><sup>4</sup> and 10<sup>−</sup><sup>5</sup> ). Meanwhile, the entire manipulated variable varies independently to control the composition indirectly.

Load change of ±20% in benzene composition of the feed was also given to check the robustness of ANNPC as shown in **Figure 8**. In this case also, the controller was able to bring the temperatures in all the sections to the desired set points without any offset.

#### **7.2 Effect of toluene composition change in feed**

**Figure 9** shows the effect of ±10% change in toluene composition of the feed as a load change in the system. Temperature overshoot in the section 1 at +10% change is 0.81°C and at −10% change is 0.41°C. The rest two sections have overshoot of 0.01 and 0.05°C, respectively. The temperatures acquire the desired set point without showing any offset in sections 1, 3, and 4 up to 0.8, 0.7 and 0.5 h, respectively. Corresponding to temperature variation in the individual section, composition also varies. At ±10% change in toluene composition, benzene composition shows the maximum peak in comparison to the toluene and xylene compositions. Due to high peak in the benzene composition, the settling time is also more (i.e., 0.66 h) with respect to the other two composition settling times. All the compositions in the different sections are also controlled with marginal offset. Load change of ±20% in

**65**

**Figure 9.**

**Figure 8.**

*±20% change in benzene composition in feed.*

*Artificial Intelligent-Based Predictive Control of Divided Wall Column*

toluene composition of the feed was also given to check the robustness of ANNPC as shown in **Figure 10**. In this case also, the controller was able to bring the tem-

A load change of ±10% in o-xylene composition was also imposed to check the performance of the controller as shown in **Figure 11**. As a ±10% change in o-xylene composition, temperature overshoot in the last section is near to 0.06°C. In comparison to the stripping section, rectifying section shows the maximum peak and

peratures in all the sections to the desired set points without any offset.

**7.3 Effect of o-xylene composition change in feed**

*±10% change in toluene composition in feed.*

*DOI: http://dx.doi.org/10.5772/intechopen.81261*

**Figure 7.** *±10% change in benzene composition in feed.*

*Artificial Intelligent-Based Predictive Control of Divided Wall Column DOI: http://dx.doi.org/10.5772/intechopen.81261*

**Figure 8.** *±20% change in benzene composition in feed.*

*Applied Modern Control*

offset and small peaks (i.e., in the range of 10<sup>−</sup><sup>4</sup>

**7.2 Effect of toluene composition change in feed**

10<sup>−</sup><sup>1</sup>

any offset.

the temperatures in all the sections to the desired set points without any offset. The settling time in section 1 and section 4 for ±10% change in benzene composition is nearly 0.6 and 0.45 h, respectively. Although main column has 0.32 and 0.42 h settling time for −10% change and +10% change in benzene composition, respectively. The maximum peak of temperature in the section one (Tsec1) is 0.4 and 0.28°C at −10% change and +10% change, respectively. Section 3 temperature (Tsec3) and section 4 temperature (Tsec4) have very small peak value in the range of 10<sup>−</sup><sup>3</sup>

°C for the ±10% change. With respect to the temperature change of the section 1, benzene composition reaches to the desired set point without any offset in 0.6 and 0.7 h for −10% and +10% change. Moreover, toluene and xylene composition acquires the desired value in 0.5 and 0.4 h, respectively without showing moderate

manipulated variable varies independently to control the composition indirectly. Load change of ±20% in benzene composition of the feed was also given to check the robustness of ANNPC as shown in **Figure 8**. In this case also, the controller was able to bring the temperatures in all the sections to the desired set points without

**Figure 9** shows the effect of ±10% change in toluene composition of the feed as a load change in the system. Temperature overshoot in the section 1 at +10% change is 0.81°C and at −10% change is 0.41°C. The rest two sections have overshoot of 0.01 and 0.05°C, respectively. The temperatures acquire the desired set point without showing any offset in sections 1, 3, and 4 up to 0.8, 0.7 and 0.5 h, respectively. Corresponding to temperature variation in the individual section, composition also varies. At ±10% change in toluene composition, benzene composition shows the maximum peak in comparison to the toluene and xylene compositions. Due to high peak in the benzene composition, the settling time is also more (i.e., 0.66 h) with respect to the other two composition settling times. All the compositions in the different sections are also controlled with marginal offset. Load change of ±20% in

and 10<sup>−</sup><sup>5</sup>

and

). Meanwhile, the entire

**64**

**Figure 7.**

*±10% change in benzene composition in feed.*

**Figure 9.** *±10% change in toluene composition in feed.*

toluene composition of the feed was also given to check the robustness of ANNPC as shown in **Figure 10**. In this case also, the controller was able to bring the temperatures in all the sections to the desired set points without any offset.

#### **7.3 Effect of o-xylene composition change in feed**

A load change of ±10% in o-xylene composition was also imposed to check the performance of the controller as shown in **Figure 11**. As a ±10% change in o-xylene composition, temperature overshoot in the last section is near to 0.06°C. In comparison to the stripping section, rectifying section shows the maximum peak and

**Figure 10.** *±20% change in toluene composition in feed.*

**Figure 11.**

*±10% change in xylene composition in feed.*

the main column shows minimum peak. After giving a load change of ±10% in o-xylene composition, benzene composition gets settled in 0.32 h, while toluene and o-xylene compositions take only 0.22 h to achieve the steady-state condition. Moreover, all the composition peaks are in the range of 1 × 10<sup>−</sup><sup>3</sup> (mole fraction), which is very small. Settling time of the temperature profile is 0.5 h in the rectifying section; 0.3 h in the main column and stripping section. Composition profiles of the sections 3 and 4 show very small offset and section 1 does not have any offset. The controller also showed the robustness for ±20% change in xylene composition in the feed as shown in **Figure 12**.

**67**

**Figure 13.**

*±10% change in feed flow rate.*

*Artificial Intelligent-Based Predictive Control of Divided Wall Column*

About ±10% load change was given in the feed flow rate to see the performance

of ANNPC as given in **Figure 13**. The temperatures in all the three sections are brought back to the set point without significant offset. The temperature overshoot is very high in section 1 at +10% change in feed flow rate in comparison to −10% change. Moreover, sections 3 and four have similar peaks in both sides. Due to the high temperature peak in section 1 at +10% change, the settling time is twice of that

*DOI: http://dx.doi.org/10.5772/intechopen.81261*

**7.4 Load change in feed low rate**

*±20% change in xylene composition in feed.*

in −10% change.

**Figure 12.**

*Artificial Intelligent-Based Predictive Control of Divided Wall Column DOI: http://dx.doi.org/10.5772/intechopen.81261*

**Figure 12.** *±20% change in xylene composition in feed.*

*Applied Modern Control*

**Figure 10.**

*±20% change in toluene composition in feed.*

**66**

**Figure 11.**

the feed as shown in **Figure 12**.

*±10% change in xylene composition in feed.*

the main column shows minimum peak. After giving a load change of ±10% in o-xylene composition, benzene composition gets settled in 0.32 h, while toluene and o-xylene compositions take only 0.22 h to achieve the steady-state condition.

which is very small. Settling time of the temperature profile is 0.5 h in the rectifying section; 0.3 h in the main column and stripping section. Composition profiles of the sections 3 and 4 show very small offset and section 1 does not have any offset. The controller also showed the robustness for ±20% change in xylene composition in

(mole fraction),

Moreover, all the composition peaks are in the range of 1 × 10<sup>−</sup><sup>3</sup>

#### **7.4 Load change in feed low rate**

About ±10% load change was given in the feed flow rate to see the performance of ANNPC as given in **Figure 13**. The temperatures in all the three sections are brought back to the set point without significant offset. The temperature overshoot is very high in section 1 at +10% change in feed flow rate in comparison to −10% change. Moreover, sections 3 and four have similar peaks in both sides. Due to the high temperature peak in section 1 at +10% change, the settling time is twice of that in −10% change.

**Figure 13.** *±10% change in feed flow rate.*

**Figure 14.** *±10% change in liquid split faction.*

Change in feed flow rate creates more disturbances in the main column of the DWC, and therefore, the settling time in this section is more in comparison to the others. The peak of benzene composition at +10% change is 0.0346°C and −10% change is 5.4 × 10<sup>−</sup><sup>3</sup> °C, which is too low as compared to +10% change in feed flow rate.

#### **7.5 Load change in liquid split factor**

To analyze the performance of the ANNPC, liquid split factor was also considered as load disturbance as shown in **Figure 14**. Due to liquid distribution in the main column and the prefractionator column, variation in temperature profile is more. Settling time in section 1 is more in comparison to the other load changes. The overshoot in this section is 0.017 (mole fraction) at +10% change and 0.019 (mole fraction) at −10% change. Overshoot in the section 3 is 0.016 (mole fraction) only at −10% change. Moreover, o-xylene composition overshoot in section 4 is in the range of 1 × 10<sup>−</sup><sup>3</sup> (mole fraction).

#### **8. Conclusion**

To analyze the control performance of artificial neural network, random training data were generated. In total, 700 data were generated on the time interval of 100 s. Generated data were divided into the ratio of 70:15:15 for training, testing, and validation, respectively. Architecture of the neural network was assumed to consist of 90 input and 3 output neurons, which showed R2 of 96% for validation. The load changes in feed flow rate, feed composition, and liquid split factor were also induced to find the control performance in all the three sections by manipulating variables, viz. reflux rate, side stream flow rate, and reboiler heat duty. Settling time in the ANNPC controller was found to be very low in comparison to conventional controller. The ANNPC was also tested for ±20% change in feed composition of benzene, toluene, and xylene, which confirmed the robustness of this controller with respect to change in feed composition.

**69**

**Author details**

Rajeev Kumar Dohare

provided the original work is properly cited.

Malaviya National Institute of Technology, Jaipur, India

\*Address all correspondence to: rajeevdohare@gmail.com

© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium,

*Artificial Intelligent-Based Predictive Control of Divided Wall Column*

*DOI: http://dx.doi.org/10.5772/intechopen.81261*

*Artificial Intelligent-Based Predictive Control of Divided Wall Column DOI: http://dx.doi.org/10.5772/intechopen.81261*

*Applied Modern Control*

change is 5.4 × 10<sup>−</sup><sup>3</sup>

*±10% change in liquid split faction.*

**Figure 14.**

the range of 1 × 10<sup>−</sup><sup>3</sup>

**8. Conclusion**

**7.5 Load change in liquid split factor**

(mole fraction).

with respect to change in feed composition.

Change in feed flow rate creates more disturbances in the main column of the DWC, and therefore, the settling time in this section is more in comparison to the others. The peak of benzene composition at +10% change is 0.0346°C and −10%

To analyze the performance of the ANNPC, liquid split factor was also considered as load disturbance as shown in **Figure 14**. Due to liquid distribution in the main column and the prefractionator column, variation in temperature profile is more. Settling time in section 1 is more in comparison to the other load changes. The overshoot in this section is 0.017 (mole fraction) at +10% change and 0.019 (mole fraction) at −10% change. Overshoot in the section 3 is 0.016 (mole fraction) only at −10% change. Moreover, o-xylene composition overshoot in section 4 is in

To analyze the control performance of artificial neural network, random training data were generated. In total, 700 data were generated on the time interval of 100 s. Generated data were divided into the ratio of 70:15:15 for training, testing, and validation, respectively. Architecture of the neural network was assumed to consist of 90 input and 3 output neurons, which showed R2 of 96% for validation. The load changes in feed flow rate, feed composition, and liquid split factor were also induced to find the control performance in all the three sections by manipulating variables, viz. reflux rate, side stream flow rate, and reboiler heat duty. Settling time in the ANNPC controller was found to be very low in comparison to conventional controller. The ANNPC was also tested for ±20% change in feed composition of benzene, toluene, and xylene, which confirmed the robustness of this controller

°C, which is too low as compared to +10% change in feed flow rate.

**68**

#### **Author details**

Rajeev Kumar Dohare Malaviya National Institute of Technology, Jaipur, India

\*Address all correspondence to: rajeevdohare@gmail.com

© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

#### **References**

[1] Willis MJ, Di Massimo C, Montague GA, Tham MT, Morris AJ. Artificial neural networks in process engineering. IEE Proceedings D Control Theory and Applications. IET. 1991;**138**(3):256-266

[2] Dohare RK, Singh K, Kumar R, Upadhyaya SG, Gupta S. Dynamic model of dividing wall column for separation of ternary system. In: Chemca-2011; Australia. 2011

[3] Dohare R, Singh K, Kumar R. Modeling and model predictive control of dividing wall column for separation of BTX. Systems Science & Control Engineering. 2015;**3**(1):142-115

[4] Park S, Han C. A nonlinear soft sensor based on multivariate smoothing procedure for quality estimation in distillation columns. Computers & Chemical Engineering. 2000;**24**:871-877

Section 2

Control Techniques

71

[5] Delgado A, Kambhampati C, Warwick K. Dynamic recurrent neural network for system identification and control. IEE Proceedings-Control Theory and Applications. 1995;**142**:307-314

[6] Nogaard M, Ravn OE, Poulsen NK, Hansen LK. Neural Networks for Modelling and Control of Dynamic Systems: A Practitioner's Handbook. London: Springer-Verlag, Inc; 2000
