Preface

Evolution of global technologies has prompted increasing complexity of applications developed in both, the industry and the scientific research fields. These complexities are generally attributed to nonlinearities, poorly defined dynamics and absence of *apriori* information about the systems. Imprecision, uncertainties and vagueness in information about the system are also playing vital roles in enhancing the complexity of application. During the last five decades researchers had concentrated their efforts on providing simple and easy algorithms using different methodologies to cope with the increasing complexity of the system. The concept of fuzzy sets had been proposed in 1965 to address the issues of application complexities arising due to nonlinearities, poorly defined dynamics, absence of *apriori* information, imprecision, uncertainties and vague description of the system. After the proposition of fuzzy set theory, this field has witnessed an explosion of its application in diverse multidiscipline areas like engineering, medicine, management, behavioral science etc. The application of fuzzy sets has widen the horizon of technologies, such as fuzzification of pixel intensity values and fuzzy clustering on image processing, fuzzy clustering on classification, decision making, identification and fault detection, fuzzy controllers to map expert knowledge into control systems, fuzzy modeling combining expert knowledge, fuzzy optimization to solve design problems etc. A very interesting characteristic of the fuzzy systems is their capability to handle numeric and linguistic information in the same framework. This characteristic made these systems very useful to handle expert tasks. Making use of these characteristic fuzzy systems are applied to artificial intelligence for representing the knowledge of experts or acquired through learning process. Additionally, fuzzy systems provide a rich and robust method of building systems that include multiple conflicting, cooperating, and collaborating knowledge.

While several books are available today that address the mathematical and philosophical foundations specific to a particular application of fuzzy logic, none, unfortunately, provides the new practitioner, with capabilities and practical information about diverse inter-disciplinary fuzzy system applications. This book is an attempt to accrue the researches performed by the prominent researchers, geographically scattered on the globe, on diverse inter disciplinary field of engineering and management using Fuzzy Inference System (FIS). The book is spread over twenty two chapters carved up in seven sections covering a wide range of applications.

#### X Preface

Section I, consists of single chapter that caters theoretical aspects of FIS. Chapter 1 deals with an introductive study about Fuzzy Logic, its differences with the other many-valued calculi and relationships with the complex sciences.

Preface XI

analysis system embedding neuron-fuzzy prediction scheme together with support vector machine in feature extraction for brain-computer interface (BCI) application.

Section V composed of chapters 15 to 17 to describe FIS application to power system engineering problem. Chapter 15 proposes a novel ANFIS based method for identifying various fault types, fault location and power restoration plan in power distribution network and evaluates the performance of proposed approach on a 47 buses practical system. Chapter 16 investigates the use of multi ANFIS to study the design of Short-Term Load Forecasting (STLF) systems for the east of Iran. The results show that temperature and the features of 2, 7 and 14 day ago have an important role in load forecast. Chapter 17 analyses the use of FIS in modeling the energy demand, improving the prediction performance and adapting the prediction to the real time

effects in a specific electric network after analyzing its demand characteristics.

Section VI constitutes chapters 18 to 20 for highlighting the FIS application to system modeling and control problems. Chapter 18 explains the three-step structure designing of fuzzy knowledge based controllers, i.e., fuzzification, inference and defuzzification using a trivial, academic example involving temperature control. Chapter 19 presents an evolutionary fuzzy hybrid system for control application by proposing the strategies for membership function generation using three different evolutionary algorithms namely modified genetic algorithms (MGA), particle swarm optimization (PSO) and hybrid particle swarm optimization (HPSO). Performance of these three different evolutionary algorithms are evaluated and compared. Chapter 20 proposes self-constructing Fuzzy Cerebellar Model Articulation Controllers (SC-FCMAC) and parametric FCMAC (P-FCMAC) as improvements in FCMAC, which demonstrates state-of-the art in the field of fuzzy inference systems for system modeling and control.

Section VII accommodates chapters 21 and 22 as FIS application to civil engineering problem. Chapter 21 proposes FIS based Bridge Management System (BMS) comprised of a Diagnosis Synthesis (DIASYN) tool that is a fuzzy rule-based inference system for bridge damage diagnosis and prediction, an adaptive neuro-fuzzy inference system for bridge risk assessment, a neuro-fuzzy hybrid system for condition state evaluation of existing reinforced concrete bridges, a fuzzy concrete bridge deck condition rating method and a two stage method for structural damage identification using hybrid of ANFIS and practice swarm optimization (PSO). Chapter 22 propounds ANFIS and Artificial Neural Network (ANN) models for prediction of shear strength of ferro-cement members and concrete beams reinforced with fiber reinforced polymer (FRP) bars. The results of these models have been compared with experimental and available methods results. The comparison shows that ANFIS and ANN have the ability to predict the shear strength of ferro-cement members and the shear strength of

**Professor (Dr.) Mohammad Fazle Azeem,**

PA College of Engineering,

India

Department of Electronics and Communication Engineering,

concrete beams reinforced with FRP with a high degree of accuracy.

Section II deals FIS applications to management related problems by enveloping chapters 2 to 4. Chapter 2 describes the main features a fuzzy expert system for supporting risk management activities termed as RA\_X FMADM (Fuzzy Multiple Attribute Decision Making) model. The support of a proactive risk management is achieved by assessing potential factors that contribute for occupational accident occurrence and by guiding on the adoption of safety measures. In its current stage a prototype was implemented for test and validation purposes. Chapter 3 introduces a new fuzzy approach to perform a more applicable risk analysis in real world applications. The introduced method is applied to two different problems. The first application is to determine the multi-purpose criticalities of activities. Whereas, second application deal with simultaneous task scheduling and path planning of rescue robots. Chapter 4 explores the potentials of the hybrid approach with the FIS to handle uncertainties in the decision making obtained from the human experts. To reduce the size of complete fuzzy rule base, because of large number of fuzzy term sets and antecedents, a rough–fuzzy approach is adopted in this chapter to form a concise fuzzy rule base with an application to reason the student performance.

Section III accumulates chapters 5 to 10 to commemorate FIS application to mechanical and industrial engineering problems. Chapter 5 presents Adaptive Neuro-Fuzzy Inference System (ANFIS) based modeling and various grip force control schemes of an intelligent robotic gripper. These schemes have been tested using simulation studies and obtained results were compared. Chapter 6 accommodates various fuzzy logic control schemes for mechatronics and automation problem like quarter car suspension system, rotary crane system automation, point to point position control, mobile autonomous robot system. Chapter 7 describes a FIS based method for the analysis of vibrations in electric machines. Chapter 8 proposes a hybrid intelligent method based on FIS to diagnose incipient and compound faults of large-scale and complex mechanical equipments like rotating machinery. Chapter 9 suggests studies on noise and its effects on industrial cognitive task performance by developing a neuro-fuzzy model for the prediction of cognitive task efficiency as a function of noise level, cognitive task type and age. Chapter 10 presents practical applications of FIS based data processing for detection of rare data in industrial applications that are capable to outperform the widely adopted traditional methods.

Section IV encompasses chapters 11 to 14 and elaborates FIS application to image processing and cognition problems. Chapter 11 describes application of FIS to image classification in the industrial field through several examples to demonstrate applicability of the FISs in industrial field because of their flexibility and the simplicity. Chapter 12 explores the superiority of FIS over the Sobel and Laplacian-of-Gaussian (LoG) operators for edge detection on some gray images. Chapter 13 presents Type-2 FIS for edge detection of gray scale images and compares the result with Type-1 FIS and magnitude gradient methods. Chapter 14 proposes a robust analysis system embedding neuron-fuzzy prediction scheme together with support vector machine in feature extraction for brain-computer interface (BCI) application.

X Preface

Section I, consists of single chapter that caters theoretical aspects of FIS. Chapter 1 deals with an introductive study about Fuzzy Logic, its differences with the other

Section II deals FIS applications to management related problems by enveloping chapters 2 to 4. Chapter 2 describes the main features a fuzzy expert system for supporting risk management activities termed as RA\_X FMADM (Fuzzy Multiple Attribute Decision Making) model. The support of a proactive risk management is achieved by assessing potential factors that contribute for occupational accident occurrence and by guiding on the adoption of safety measures. In its current stage a prototype was implemented for test and validation purposes. Chapter 3 introduces a new fuzzy approach to perform a more applicable risk analysis in real world applications. The introduced method is applied to two different problems. The first application is to determine the multi-purpose criticalities of activities. Whereas, second application deal with simultaneous task scheduling and path planning of rescue robots. Chapter 4 explores the potentials of the hybrid approach with the FIS to handle uncertainties in the decision making obtained from the human experts. To reduce the size of complete fuzzy rule base, because of large number of fuzzy term sets and antecedents, a rough–fuzzy approach is adopted in this chapter to form a concise

many-valued calculi and relationships with the complex sciences.

fuzzy rule base with an application to reason the student performance.

capable to outperform the widely adopted traditional methods.

Section III accumulates chapters 5 to 10 to commemorate FIS application to mechanical and industrial engineering problems. Chapter 5 presents Adaptive Neuro-Fuzzy Inference System (ANFIS) based modeling and various grip force control schemes of an intelligent robotic gripper. These schemes have been tested using simulation studies and obtained results were compared. Chapter 6 accommodates various fuzzy logic control schemes for mechatronics and automation problem like quarter car suspension system, rotary crane system automation, point to point position control, mobile autonomous robot system. Chapter 7 describes a FIS based method for the analysis of vibrations in electric machines. Chapter 8 proposes a hybrid intelligent method based on FIS to diagnose incipient and compound faults of large-scale and complex mechanical equipments like rotating machinery. Chapter 9 suggests studies on noise and its effects on industrial cognitive task performance by developing a neuro-fuzzy model for the prediction of cognitive task efficiency as a function of noise level, cognitive task type and age. Chapter 10 presents practical applications of FIS based data processing for detection of rare data in industrial applications that are

Section IV encompasses chapters 11 to 14 and elaborates FIS application to image processing and cognition problems. Chapter 11 describes application of FIS to image classification in the industrial field through several examples to demonstrate applicability of the FISs in industrial field because of their flexibility and the simplicity. Chapter 12 explores the superiority of FIS over the Sobel and Laplacian-of-Gaussian (LoG) operators for edge detection on some gray images. Chapter 13 presents Type-2 FIS for edge detection of gray scale images and compares the result with Type-1 FIS and magnitude gradient methods. Chapter 14 proposes a robust Section V composed of chapters 15 to 17 to describe FIS application to power system engineering problem. Chapter 15 proposes a novel ANFIS based method for identifying various fault types, fault location and power restoration plan in power distribution network and evaluates the performance of proposed approach on a 47 buses practical system. Chapter 16 investigates the use of multi ANFIS to study the design of Short-Term Load Forecasting (STLF) systems for the east of Iran. The results show that temperature and the features of 2, 7 and 14 day ago have an important role in load forecast. Chapter 17 analyses the use of FIS in modeling the energy demand, improving the prediction performance and adapting the prediction to the real time effects in a specific electric network after analyzing its demand characteristics.

Section VI constitutes chapters 18 to 20 for highlighting the FIS application to system modeling and control problems. Chapter 18 explains the three-step structure designing of fuzzy knowledge based controllers, i.e., fuzzification, inference and defuzzification using a trivial, academic example involving temperature control. Chapter 19 presents an evolutionary fuzzy hybrid system for control application by proposing the strategies for membership function generation using three different evolutionary algorithms namely modified genetic algorithms (MGA), particle swarm optimization (PSO) and hybrid particle swarm optimization (HPSO). Performance of these three different evolutionary algorithms are evaluated and compared. Chapter 20 proposes self-constructing Fuzzy Cerebellar Model Articulation Controllers (SC-FCMAC) and parametric FCMAC (P-FCMAC) as improvements in FCMAC, which demonstrates state-of-the art in the field of fuzzy inference systems for system modeling and control.

Section VII accommodates chapters 21 and 22 as FIS application to civil engineering problem. Chapter 21 proposes FIS based Bridge Management System (BMS) comprised of a Diagnosis Synthesis (DIASYN) tool that is a fuzzy rule-based inference system for bridge damage diagnosis and prediction, an adaptive neuro-fuzzy inference system for bridge risk assessment, a neuro-fuzzy hybrid system for condition state evaluation of existing reinforced concrete bridges, a fuzzy concrete bridge deck condition rating method and a two stage method for structural damage identification using hybrid of ANFIS and practice swarm optimization (PSO). Chapter 22 propounds ANFIS and Artificial Neural Network (ANN) models for prediction of shear strength of ferro-cement members and concrete beams reinforced with fiber reinforced polymer (FRP) bars. The results of these models have been compared with experimental and available methods results. The comparison shows that ANFIS and ANN have the ability to predict the shear strength of ferro-cement members and the shear strength of concrete beams reinforced with FRP with a high degree of accuracy.

> **Professor (Dr.) Mohammad Fazle Azeem,** Department of Electronics and Communication Engineering, PA College of Engineering, India

**Section 1** 

**Theory** 
