**3.3 Process analysis and optimization**

The applications of neural networks to the process analysis are increasing. Assidjo et al. [45] modeled the drying process of the production of coconut using a neural network. The goal is to predict the moisture of dried grated coconut whose dynamics are not well known. The authors used a feedforward fully connected neural network, whereby the selected architecture was 9–4-1, selected based on the minimum error in the test set. The results indicate that the neural network proposed, constructed using industrial plant data, can be used as a predicting method.

Fernandes and Lona [46] applied neural networks to the field of polymerization. The authors also highlighted some topologies, the number of data points needed, and the concept of stacked neural networks that can enhance the prediction of the final model.

Alves and Nascimento [47] used industrial plant data for constructing neural networks to detect gross errors; the case study was an isoprene unit facility.

Alves and Nascimento [4] studied the production of high purity isoprene from a C5 cut arising from a pyrolysis gasoline unit. The first principle models were replaced by neural networks in the final grid search of the optimal parameters for

**173**

**Table 2.**

*\**

*Application of Artificial Neural Networks to Chemical and Process Engineering*

Catalytic activity for n-paraffin isomerization

Design of catalyst for propane ammoxidation

> Design of a catalyst for methane oxidative coupling

Analysis of NO decomposition over Cu/ZSM-5 zeolite

Modeling of catalysts for oxidative dehydrogenation of ethane

Estimation of the reaction rate in methanol dehydration

Selective CO Oxidation over Copper-Based Catalysts

Catalyst selection for the WGS reaction

Fischer-Tropsch synthesis to lower-olefins

Dry reformer under catalyst sintering

Determination of acidity in metal incorporated zeolites by FTRI

*\*\*The first and last elements in topology represent the number of neurons in the input and in the output layer,* 

under study using neural networks with industrial data.

**References Field Case Study Class of** 

the process. The set of 10 neural networks were defined to represent the whole flowsheeting, whereby the number of hidden layers was defined by the minimum error in the test set. Lastly, the framework successfully optimized a chemical plant

> **Neural Network**

FF-ANN\* Sigmoid/

FF-ANN Not

FF-ANN Tanh/

described

Linear

**Activation Function**

Tanh

FF-ANN Sigmoid 6–20–12–2 in-house

FF-ANN Sigmoid 6–20–9-2 in-house

FF-ANN Sigmoid 4–32-1 in-house

FF-ANN Tanh 14–7–7-1 Matlab

FF-ANN Sigmoid 51–12-1 R - neuralnet

FF-ANN Sigmoid 30–15-2 R - neuralnet

FF-ANN Tanh 3–12–5-6-1 in-house

FF-ANN Tanh 6–10-1 Matlab

**Topology\*\* Software**

4–8–6-3 SNNS neural

networks simulator

software

software

software

networks simulator

software

13–26–12-6 SNNS neural

3–6-1 Matlab

*DOI: http://dx.doi.org/10.5772/intechopen.96641*

[29] Modeling

[32] Catalyst

[33] Catalyst

[34] Modeling

[36] Combinatorial

[37] Modeling

[38] Modeling

[8] Catalyst

[41] Modeling

[42] Catalyst

[43] Determination

of catalytic processes

design

design

of catalytic processes

catalysis

of catalytic processes

of catalytic processes

selection

of catalytic processes

deactivation

of catalyst acidity

*Current applications of ANNs to catalytic processes.*

*FN-ANN stands for Feed-Forward Artificial Neural Network.*

*respectively. Among them, the number of neurons in the hidden layer(s).*

the process. The set of 10 neural networks were defined to represent the whole flowsheeting, whereby the number of hidden layers was defined by the minimum error in the test set. Lastly, the framework successfully optimized a chemical plant under study using neural networks with industrial data.


*\* FN-ANN stands for Feed-Forward Artificial Neural Network.*

*\*\*The first and last elements in topology represent the number of neurons in the input and in the output layer, respectively. Among them, the number of neurons in the hidden layer(s).*

#### **Table 2.**

*Deep Learning Applications*

hydrogen - a clean energy source.

that performed well over testing data points.

**3.3 Process analysis and optimization**

and operating conditions.

ideal composition of the catalyst in the water-gas-shift reaction and discover useful trends through sensitivity analysis. The input variables for ANN were several, while the only output variable considered was the conversion of CO. The model for the reaction was successfully developed, exhibiting the power of ANNs for predicting

Recently, Cavalcanti et al. [8] showed that ANNs are able to predict the variables

In the same topic, Garona et al. [41] presented an empiric model for the Fischer-Tropsch Synthesis (FTS) reaction using ANNs. A database of FTS to light olefins was assembled from the literature, and feedforward neural networks were used to build more complete models, which helped to predict optimal catalyst composition

Another application is in the determination of acidity in zeolites with data from FTIR spectroscopy [43]. FF-ANNs were used for analyzing multivariate base on the characteristic absorbance of 11 zeolite samples after metal substitution (Zn, Cu, Ga, and Ag) in the ~3612 cm−1 region. The developed regression method presented the same results of acid sites from other conventional and expensive methodologies. Thus, in order to formulate a new kind of catalyst, it is essential to identify the catalysis past [44]. Therefore, by using ANNs, it is possible to convert historical data from past publications into valuable information, leading to a great acceleration in the development of new catalysts with better performances for a given process [8]. **Table 2** presents

a summary of the current applications of neural networks to catalytic processes.

The applications of neural networks to the process analysis are increasing. Assidjo et al. [45] modeled the drying process of the production of coconut using a neural network. The goal is to predict the moisture of dried grated coconut whose dynamics are not well known. The authors used a feedforward fully connected neural network, whereby the selected architecture was 9–4-1, selected based on the minimum error in the test set. The results indicate that the neural network proposed, constructed using industrial plant data, can be used as a predicting method. Fernandes and Lona [46] applied neural networks to the field of polymerization. The authors also highlighted some topologies, the number of data points needed, and the concept of stacked neural networks that can enhance the prediction of the

Alves and Nascimento [47] used industrial plant data for constructing neural

Alves and Nascimento [4] studied the production of high purity isoprene from

networks to detect gross errors; the case study was an isoprene unit facility.

a C5 cut arising from a pyrolysis gasoline unit. The first principle models were replaced by neural networks in the final grid search of the optimal parameters for

It is also noteworthy that ANNs were also used to model the sintering of a catalyst in a dry reformer [42]. In particular, the effects of temperature, pressure, and catalyst diameter on methane and CO2 conversions, H2/CO ratio, and molar percentage of solid carbon deposited on the catalyst (responsible for deactivation) have been studied. The ANN design activity was automated using a Genetic Algorithm (GA) search over the set of possible network topologies. The inclusion of the effective number of parameters in the GA objective function led to networks

that most influence the conversion of CO in the water-gas-shift reaction, that is, temperature and surface area. The results can be used to conduct subsequent research in an optimized manner in this area, as it aims at the well-managed use of environmental resources, in the sense of selecting efficient catalysts for producing

better catalysts and operating conditions for the process.

**172**

final model.

*Current applications of ANNs to catalytic processes.*

#### *Deep Learning Applications*

Khezri et al. [15] proposed a hybrid model for optimizing a large-scale gas to liquids process. The dataset was constructed using a simulation model of the GTL process. Different topologies were compared to select the most promising one; one and two hidden layers with different number of neurons were tested. The optimal configuration was two hidden layers with 7 and 15 hidden neurons each. The ANN was modeled using the information of the tail gas unpurged ratio, recycled tail gas to FT ratio, H2O/C in the syngas section, and CO2 removal percentage as input features; the outputting was wax production rate. The ANN model was then used for optimization purposes.

Wang et al. [16] proposed a framework for predicting the operating trend of an industrial process. The framework contains three major steps: (i) multivariate correlation analysis, to deal with the correlation between the historical industrial data, (ii) clustering, due to nonlinear dense data and unclear operating trend types and (iii) a convolutional neural network (CNN), formed by five parts (input layer, convolutional layer, ReLU layer, pooling layer, and fully connected layer).

The authors pointed out the importance of the convolutional networks to extract important features from the dataset. Moreover, the advantage of such a framework was compared with traditional convolutional neural networks and recurrent neural networks (RNNs) for a methanol production process.

Cai et al. [48] analyzed an industrial process using data-driven models. The case study was the industrial reverse osmosis concentrate (ROC) treatment with the fluidized bed reactor Fenton (FBR-Fenton) process. Prior to modeling, a statistical analysis was carried out to determine the most relevant features as input (Fe2+ dosage, H2O2 dosage, pH, and HRT). Two approaches were studied, ANN and linear regression. The former showed more accurate predictions, consisting in one input layer (4 neurons), 4 hidden layers (10 neurons each) and one output layer (2 neurons) using ReLU as an activation function, due to the least computationally dense mechanism and also a general approximation for most scenarios [11].

The crystallization process and the quality of the products was studied by Lin et al. [49]. The authors used a Raman spectrum as input for a two-layer back propagation neural network with four hidden neurons to predict the solution concentration and slurry density simultaneously. They also compared the output prediction of the neural network with other algorithm predictions (characteristic peaks regression, principal component regression, partial least-squares regression), and the results indicated the superior prediction characteristics of the neural network due to its inner nonlinear nature.

Chemical process synthesis is a complex scheme, which comprises process modeling and design, and combinatorial defiance. There are two major approaches: the traditional sequential form and the optimization-based synthesis using superstructure models. In the former category, the problem is solved in sequential scheme, by decomposition whereby there is a hierarchy of elements that can be depicted by an Onion Diagram (reactor, separation, heat recovery and utility) [50].

The latter category considers the full integration between decisions at the single step, i.e. determine the optimal structure and operating conditions simultaneously. Therefore, this approach contemplates all possible complex interactions between the engineering choices, including equipment (potentially selected in the optimized flowsheet), the interconnection and operating conditions formulated as an optimization problem [51–53].

There is a diversity of proposed methodologies to represent a general process superstructure [54–56]. However, due to the inner complexity of the superstructure (**Figure 1**), the large-scale non-convex Mixed-Integer Nonlinear Programs (MINLP) require effective approaches to solve them.

**175**

*Application of Artificial Neural Networks to Chemical and Process Engineering*

*Simple superstructure representation compared with different separation processes.*

The use of simplified models or surrogates at the unit operation level is advantageous because they are present in any process simulator. Additionally, surrogates can be used to represent an entire subsystem consisting of a definite number of units. Artificial Neural Networks (ANNs) may be used to generate the surrogate

In order to circumvent the solution problem of a superstructure, Henao and Maravelias [58] proposed a framework to replace complex unit models (based on first-principle) with surrogate models, developed using artificial neural networks. The authors proposed simpler surrogate models for pumps, compressors and flash vessels. The authors used two case studies (Absorption-based CO2 capture system and maleic anhydride process superstructure) to validate the proposed framework. The results indicate the possibility of using neural networks embedded in a rigorous

Savage et al. [59] proposed a hybrid machine learning-based framework to optimize the chemical process (the CryoMan Cascade cycle system was used as a case study). The authors compared different surrogate models algorithms (ANN and Kriging Partial Least Squares); the results indicated a reduction in the time needed for the optimization when compared with the rigorous model. Moreover, they found that a single large ANN was unable to capture the high nonlinearity of the process under study based on the final accuracy. Therefore, the authors broke the surrogate model into a series of parallel sub-models, revealing to have increased

According to Klemes et al. [60], despite the substantial level of maturity of the process modeling, the nature of connections of the problem still allows improvements. Nascimento et al. [61] also presented alternatives for the optimization of industrial facilities using neural networks and compared them with indus-

**Table 3** presents a summary of the current applications of neural networks to

One of the most common applications of ANNs to the area of process safety and control is in fault detection and diagnosis. These systems are built to identify habitual process behavior and recognize atypical variations in the chemical plant that can lead to an accident [64]. Generally, deep neural networks – ANNs that contain several hidden layers – are used to extract spatial and temporal aspects of the data for this purpose [65]. Their inputs are the sensors responsible for the variable measurement, and their outputs of the kind of faults (e.g., tube plugging, valve

*DOI: http://dx.doi.org/10.5772/intechopen.96641*

models, due to their fitting characteristics [57].

optimization procedure.

**Figure 1.**

the final accuracy.

process analysis and optimization.

blockage, catalyst deactivation, among others) [66].

**3.4 Process safety and control**

trial data.

*Application of Artificial Neural Networks to Chemical and Process Engineering DOI: http://dx.doi.org/10.5772/intechopen.96641*

#### **Figure 1.**

*Deep Learning Applications*

for optimization purposes.

due to its inner nonlinear nature.

zation problem [51–53].

Khezri et al. [15] proposed a hybrid model for optimizing a large-scale gas to liquids process. The dataset was constructed using a simulation model of the GTL process. Different topologies were compared to select the most promising one; one and two hidden layers with different number of neurons were tested. The optimal configuration was two hidden layers with 7 and 15 hidden neurons each. The ANN was modeled using the information of the tail gas unpurged ratio, recycled tail gas to FT ratio, H2O/C in the syngas section, and CO2 removal percentage as input features; the outputting was wax production rate. The ANN model was then used

Wang et al. [16] proposed a framework for predicting the operating trend of an industrial process. The framework contains three major steps: (i) multivariate correlation analysis, to deal with the correlation between the historical industrial data, (ii) clustering, due to nonlinear dense data and unclear operating trend types and (iii) a convolutional neural network (CNN), formed by five parts (input layer,

The authors pointed out the importance of the convolutional networks to extract important features from the dataset. Moreover, the advantage of such a framework was compared with traditional convolutional neural networks and recurrent neural

Cai et al. [48] analyzed an industrial process using data-driven models. The case

The crystallization process and the quality of the products was studied by Lin et al. [49]. The authors used a Raman spectrum as input for a two-layer back propagation neural network with four hidden neurons to predict the solution concentration and slurry density simultaneously. They also compared the output prediction of the neural network with other algorithm predictions (characteristic peaks regression, principal component regression, partial least-squares regression), and the results indicated the superior prediction characteristics of the neural network

Chemical process synthesis is a complex scheme, which comprises process modeling and design, and combinatorial defiance. There are two major approaches: the traditional sequential form and the optimization-based synthesis using superstructure models. In the former category, the problem is solved in sequential scheme, by decomposition whereby there is a hierarchy of elements that can be depicted by an

The latter category considers the full integration between decisions at the single step, i.e. determine the optimal structure and operating conditions simultaneously. Therefore, this approach contemplates all possible complex interactions between the engineering choices, including equipment (potentially selected in the optimized flowsheet), the interconnection and operating conditions formulated as an optimi-

There is a diversity of proposed methodologies to represent a general process superstructure [54–56]. However, due to the inner complexity of the superstructure (**Figure 1**), the large-scale non-convex Mixed-Integer Nonlinear Programs

Onion Diagram (reactor, separation, heat recovery and utility) [50].

(MINLP) require effective approaches to solve them.

study was the industrial reverse osmosis concentrate (ROC) treatment with the fluidized bed reactor Fenton (FBR-Fenton) process. Prior to modeling, a statistical analysis was carried out to determine the most relevant features as input (Fe2+ dosage, H2O2 dosage, pH, and HRT). Two approaches were studied, ANN and linear regression. The former showed more accurate predictions, consisting in one input layer (4 neurons), 4 hidden layers (10 neurons each) and one output layer (2 neurons) using ReLU as an activation function, due to the least computationally dense mechanism and also a general approximation for most scenarios [11].

convolutional layer, ReLU layer, pooling layer, and fully connected layer).

networks (RNNs) for a methanol production process.

**174**

*Simple superstructure representation compared with different separation processes.*

The use of simplified models or surrogates at the unit operation level is advantageous because they are present in any process simulator. Additionally, surrogates can be used to represent an entire subsystem consisting of a definite number of units. Artificial Neural Networks (ANNs) may be used to generate the surrogate models, due to their fitting characteristics [57].

In order to circumvent the solution problem of a superstructure, Henao and Maravelias [58] proposed a framework to replace complex unit models (based on first-principle) with surrogate models, developed using artificial neural networks. The authors proposed simpler surrogate models for pumps, compressors and flash vessels. The authors used two case studies (Absorption-based CO2 capture system and maleic anhydride process superstructure) to validate the proposed framework. The results indicate the possibility of using neural networks embedded in a rigorous optimization procedure.

Savage et al. [59] proposed a hybrid machine learning-based framework to optimize the chemical process (the CryoMan Cascade cycle system was used as a case study). The authors compared different surrogate models algorithms (ANN and Kriging Partial Least Squares); the results indicated a reduction in the time needed for the optimization when compared with the rigorous model. Moreover, they found that a single large ANN was unable to capture the high nonlinearity of the process under study based on the final accuracy. Therefore, the authors broke the surrogate model into a series of parallel sub-models, revealing to have increased the final accuracy.

According to Klemes et al. [60], despite the substantial level of maturity of the process modeling, the nature of connections of the problem still allows improvements. Nascimento et al. [61] also presented alternatives for the optimization of industrial facilities using neural networks and compared them with industrial data.

**Table 3** presents a summary of the current applications of neural networks to process analysis and optimization.

#### **3.4 Process safety and control**

One of the most common applications of ANNs to the area of process safety and control is in fault detection and diagnosis. These systems are built to identify habitual process behavior and recognize atypical variations in the chemical plant that can lead to an accident [64]. Generally, deep neural networks – ANNs that contain several hidden layers – are used to extract spatial and temporal aspects of the data for this purpose [65]. Their inputs are the sensors responsible for the variable measurement, and their outputs of the kind of faults (e.g., tube plugging, valve blockage, catalyst deactivation, among others) [66].


*\* FN-ANN stands for Feed-Forward Artificial Neural Network.*

*\*\*CNN stands for Convolutional Neural Network.*

*\*\*\*RNN stands for Recurrent Neural Network.*

*\*\*\*\*The first and last elements in topology represent the number of neurons in the input and in the output layer, respectively. Among them, the number of neurons in the hidden layer(s).*

#### **Table 3.**

*Current applications of neural networks to process analysis and optimization.*

However, determining the various hyperparameters of deep neural networks demands a considerable amount of time, which is not suitable for fast online process applications. Based on this, Peng et al. [67] applied a method to reduce the training time of these complex types of network architecture: the Broad Learning

**177**

industrial plants.

*Application of Artificial Neural Networks to Chemical and Process Engineering*

System (BLS). It uses an incremental learning procedure and enlarges the network in width, making a quick training stage possible. They successfully employed this strategy in a batch fermentation process for fault detection utilizing the Affinity Propagation (AP) algorithm in a Long Short-Term Memory (LSTM) deep neural

Another use is in developing models to control the process quality through variables that do not have online sensors. On the one hand, variables such as pressure, temperature, and mass flow rate can be easily measured by manometers, thermocouples, and mass flow controllers, respectively. On the other hand, the online measurement of a variable such as pH in the process is a challenge since no large-scale equipment exists for this, depending on an offline laboratory analysis. Therefore, ANNs can be used to develop these so-called *soft-sensors* to predict quality parameters from a large volume of industrial data, improving the process control

Finally, ANNs are also used to replace complex phenomenological models in Model Predictive Control (MPC) architectures and Real-Time Optimization (RTO) strategies [69]. Both applications depend on the model accuracy and the velocity of solving the model equations to drive the controlled variable to the desired set-point. The former is related to dynamic processes and the latter to steady-state operations [69]. Since ANNs have a lower computational response than first-principle models, they are a suitable alternative to make these control strategies possible and

A successful application of this kind of substitution can be found elsewhere [70], in which an ANN is used to replace a very detailed computational fluid dynamic (CFD) model that represents the synthesis of phthalic anhydride in a fixed-bed catalytic reactor for an MPC structure. Moreover, a hybrid model approach (first-principles combined with ANN) was employed in an MPC by Zhang et al. [69] to drive a reaction process in a continuous stirred tank reactor (CSTR) to optimal operating conditions. They represented the reaction rates by neural networks instead of using the nonlinear Arrhenius Law to describe the reaction phenomenon. Indeed, this well-known equation was used to generate the dataset for training the network under numerous variations in temperature and reactant concentrations. The MPC acted to stabilize the chemical process, driving it

Wu et al. [14] proposed a hybrid machine-learning model that incorporates first principles into a recurrent neural network. The authors studied two models, a partially-connected RNN model and a weight-constrained RNN model and applied them to a chemical process containing two well-mixed, non- isothermal continuous stirred tank reactors in series. The two proposed models outperformed a Lyapunovbased model predictive controller based on prediction accuracy, smoother state

It is worth mentioning that ANNs are being used to build detectors to prevent cyber-attacks against process plants [71]. Nowadays, with highly automated systems for controlling chemical plants with real-time operation, breaches in cyber-secure failures can exist, which may cause accidents and economic losses. With this in mind, Chen et al. [71] developed a feedback-MPC control architecture with an ANN-detector that can identify the probabilities of cyber-attacks in networked sensors. Therefore, the applicability of ANNs in these safety and control strategies is very significant for the integrability of

**Table 4** shows a summary of the current applications of neural networks to the

*DOI: http://dx.doi.org/10.5772/intechopen.96641*

network to cluster distinct stage data.

to the lowest total cost conditions.

trajectories and economic advantages.

area of process safety and control.

quality [68].

efficient.

### *Application of Artificial Neural Networks to Chemical and Process Engineering DOI: http://dx.doi.org/10.5772/intechopen.96641*

System (BLS). It uses an incremental learning procedure and enlarges the network in width, making a quick training stage possible. They successfully employed this strategy in a batch fermentation process for fault detection utilizing the Affinity Propagation (AP) algorithm in a Long Short-Term Memory (LSTM) deep neural network to cluster distinct stage data.

Another use is in developing models to control the process quality through variables that do not have online sensors. On the one hand, variables such as pressure, temperature, and mass flow rate can be easily measured by manometers, thermocouples, and mass flow controllers, respectively. On the other hand, the online measurement of a variable such as pH in the process is a challenge since no large-scale equipment exists for this, depending on an offline laboratory analysis. Therefore, ANNs can be used to develop these so-called *soft-sensors* to predict quality parameters from a large volume of industrial data, improving the process control quality [68].

Finally, ANNs are also used to replace complex phenomenological models in Model Predictive Control (MPC) architectures and Real-Time Optimization (RTO) strategies [69]. Both applications depend on the model accuracy and the velocity of solving the model equations to drive the controlled variable to the desired set-point. The former is related to dynamic processes and the latter to steady-state operations [69]. Since ANNs have a lower computational response than first-principle models, they are a suitable alternative to make these control strategies possible and efficient.

A successful application of this kind of substitution can be found elsewhere [70], in which an ANN is used to replace a very detailed computational fluid dynamic (CFD) model that represents the synthesis of phthalic anhydride in a fixed-bed catalytic reactor for an MPC structure. Moreover, a hybrid model approach (first-principles combined with ANN) was employed in an MPC by Zhang et al. [69] to drive a reaction process in a continuous stirred tank reactor (CSTR) to optimal operating conditions. They represented the reaction rates by neural networks instead of using the nonlinear Arrhenius Law to describe the reaction phenomenon. Indeed, this well-known equation was used to generate the dataset for training the network under numerous variations in temperature and reactant concentrations. The MPC acted to stabilize the chemical process, driving it to the lowest total cost conditions.

Wu et al. [14] proposed a hybrid machine-learning model that incorporates first principles into a recurrent neural network. The authors studied two models, a partially-connected RNN model and a weight-constrained RNN model and applied them to a chemical process containing two well-mixed, non- isothermal continuous stirred tank reactors in series. The two proposed models outperformed a Lyapunovbased model predictive controller based on prediction accuracy, smoother state trajectories and economic advantages.

It is worth mentioning that ANNs are being used to build detectors to prevent cyber-attacks against process plants [71]. Nowadays, with highly automated systems for controlling chemical plants with real-time operation, breaches in cyber-secure failures can exist, which may cause accidents and economic losses. With this in mind, Chen et al. [71] developed a feedback-MPC control architecture with an ANN-detector that can identify the probabilities of cyber-attacks in networked sensors. Therefore, the applicability of ANNs in these safety and control strategies is very significant for the integrability of industrial plants.

**Table 4** shows a summary of the current applications of neural networks to the area of process safety and control.

*Deep Learning Applications*

[45] Process

[16] Industrial

[62] Process

[14] Predictive

[15] Process

[4] Process

[58] Process

[59] Process

[63] Process

[49] Process

Analysis

Process Operating (Predictive Control)

Analysis

Control

Optimization

Optimization

Synthesis

Synthesis

Analysis

Analysis

*\*\*CNN stands for Convolutional Neural Network. \*\*\*RNN stands for Recurrent Neural Network.*

*FN-ANN stands for Feed-Forward Artificial Neural Network.*

*respectively. Among them, the number of neurons in the hidden layer(s).*

*Current applications of neural networks to process analysis and optimization.*

**References Field Case Study Class of** 

Grated coconut industry

Methanol production

Fluidized bed reactor Fenton process

nonisothermal continuous stirred tank reactors

Large scale gas to liquids process

> Isoprene Process

Absorptionbased CO2 capture and Maleic Anhydride process

CryoMan Cascade cycle system

Thermocatalytic methane decomposition

Crystallization process

**Neural Network** **Activation Function**

CNN\*\* ReLu 5 convolution

FF-ANN ReLu 4–10–

RNN\*\*\* Tanh 2 hidden layers

FF-ANN Sigmoid 10 neural

FF-ANN Tanh Several neural

Described

Described

FF-ANN Sigmoid 6–9-1 Matlab

FF-ANN Not

FF-ANN Not

FF-ANN Sigmoid 4–7–15-1 Matlab

FF-ANN\* Tanh 9–4-1 Matlab

layers, 36 filters, and 3 pooling layers

10 − 10 − 10 − 2

with 30 neurons in each layer

networks (all with one hidden layer)

networks (all with one hidden layer)

two-layer neural network with four hidden neurons

Not Described Python-

**Topology\*\*\*\* Software**

Caffe

R - Keras

Python-Keras

in-house software

Matlab

PyTorch

Matlab

**176**

*\**

**Table 3.**

However, determining the various hyperparameters of deep neural networks demands a considerable amount of time, which is not suitable for fast online process applications. Based on this, Peng et al. [67] applied a method to reduce the training time of these complex types of network architecture: the Broad Learning

*\*\*\*\*The first and last elements in topology represent the number of neurons in the input and in the output layer,* 


*\* FN-ANN stands for Feed-Forward Artificial Neural Network.*

*\*\*RNN stands for Recurrent Neural Network.*

*\*\*\*LSTM stands for Long Short-Term Memory.*

*\*\*\*\*The first and last elements in topology represent the number of neurons in the input and in the output layer, respectively. Among them, the number of neurons in the hidden layer(s).*

#### **Table 4.**

*Current applications of ANNs to process safety and control.*
