Soft Computing and Fractal Theory for Intelligent Manufacturing

We describe in this book, new methods for intelligent manufacturing using soft computing techniques and fractal theory. Soft Computing (SC) consists of several computing paradigms, including fuzzy logic, neural networks, and genetic algorithms, which can be used to produce powerful hybrid intelligent systems. Fractal theory provides us with the mathematical tools to understand the geometrical complexity of natural objects and can be used for identification and modeling purposes. Combining SC techniques with fractal theory, we can take advantage of the "intelligence" provided by the computer methods and also take advantage of the descriptive power of the fractal mathematical tools. Industrial manufacturing systems can be considered as non-linear dynamical systems, and as a consequence can have highly complex dynamic behaviors. For this reason, the need for computational intelligence in these manufacturing systems has now been well recognized. We consider in this book the concept of "intelligent manufacturing" as the application of soft computing techniques and fractal theory for achieving the goals of manufacturing, which are production planning and control, monitoring and diagnosis of faults, and automated quality control. As a prelude, we provide a brief overview of the existing methodologies in Soft Computing. We then describe our own approach in dealing with the problems in achieving intelligent manufacturing. Our particular point of view is that to really achieve intelligent manufacturing in real-world applications we need to use SC techniques and fractal theory.

1 Introduction.- 2 Type-1 Fuzzy Logic.- 2.1 Type-1 Fuzzy Set Theory.- 2.2 Fuzzy Rules and Fuzzy Reasoning.- 2.2.1 Fuzzy Relations.- 2.2.2 Fuzzy Rules.- 2.3 Fuzzy Inference Systems.- 2.4 Fuzzy Modelling.- 2.5 Summary.- 3 Type-2 Fuzzy Logic.- 3.1 Type-2 Fuzzy Sets.- 3.2 Operations of Type-2 Fuzzy Sets.- 3.3 Type-2 Fuzzy Systems.- 3.3.1 Singleton Type-2 Fuzzy Logic Systems.- 3.3.2 Non-Singleton Fuzzy Logic Systems.- 3.3.3 Sugeno Type-2 Fuzzy Systems.- 3.4 Summary.- 4 Supervised Learning Neural Networks.- 4.1 Backpropagation for Feedforward Networks.- 4.1.1 The Backpropagation Learning Algorithm.- 4.1.2 Backpropagation Multilayer Perceptrons.- 4.1.3 Methods for Speeding up Backpropagation.- 4.2 Radial Basis Function Networks.- 4.3 Adaptive Neuro-Fuzzy Inference Systems.- 4.3.1 ANFIS Architecture.- 4.3.2 Learning Algorithm.- 4.4 Summary.- 5 Unsupervised Learning Neural Networks.- 5.1 Competitive Learning Networks.- 5.2 Kohonen Self-Organizing Networks.- 5.3 Learning Vector Quantization.- 5.4 The Hopfield Network.- 5.5 Summary.- 6 Genetic Algorithms and Simulated Annealing.- 6.1 Genetic Algorithms.- 6.2 Modifications to Genetic Algorithms.- 6.2.1 Chromosome Representation.- 6.2.2 Objective Function and Fitness.- 6.2.3 Selection Methods.- 6.2.4 Genetic Operations.- 6.2.5 Parallel Genetic Algorithm.- 6.3 Simulated Annealing.- 6.4 Applications of Genetic Algorithms.- 6.4.1 Evolving Neural Networks.- 6.4.1.1 Evolving Weights in a Fixed Network.- 6.4.1.2 Evolving Network Architectures.- 6.4.2 Evolving Fuzzy Systems.- 6.5 Summary.- 7 Dynamical Systems Theory.- 7.1 Basic Concepts of Dynamical Systems.- 7.2 Controlling Chaos.- 7.2.1 Controlling Chaos through Feedback.- 7.2.1.1 Ott-Grebogi-Yorke Method.- 7.2.1.2 Pyragas’s Control Methods.- 7.2.1.3 Controlling Chaos by Chaos.- 7.2.2 Controlling Chaos without Feedback.- 7.2.2.1 Control through Operating Conditions.- 7.2.2.2 Control by System Design.- 7.2.2.3 Taming Chaos.- 7.2.3 Method Selection.- 7.3 Summary.- 8 Plant Monitoring and Diagnostics.- 8.1 Monitoring and Diagnosis.- 8.2 Fractal Dimension of a Geometrical Object.- 8.3 Fuzzy Estimation of the Fractal Dimension.- 8.4 Plant Monitoring with Fuzzy-Fractal Approach.- 8.5 Experimental Results.- 8.6 Summary.- 9 Adaptive Control of Non-Linear Plants.- 9.1 Fundamental Adaptive Fuzzy Control Concept.- 9.2 Basic Concepts of Stepping Motors.- 9.2.1 Variable Reluctance Motors.- 9.2.2 Unipolar Motors.- 9.2.3 Bipolar Motors.- 9.2.4 Dynamics of the Stepping Motor.- 9.2.5 Control of the Stepping Motor.- 9.3 Fuzzy Logic Controller of the Stepping Motor.- 9.4 Hardware Implementation of ANFIS.- 9.5 Experimental Results.- 9.6 Summary.- 10 Automated Quality Control in Sound Speaker Manufacturing.- 10.1 Introduction.- 10.2 Basic Concepts of Sound Speakers.- 10.2.1 Sound Basics.- 10.2.2 Making Sound.- 10.2.3 Chunks of the Frequency Range.- 10.2.4 Boxes of Sound.- 10.2.5 Alternative Speaker Designs.- 10.3 Description of the Problem.- 10.4 Fractal Dimension of a Sound Signal.- 10.5 Experimental Results.- 10.6 Summary.- 11 Intelligent Manufacturing of Television Sets.- 11.1 Introduction.- 11.2 Imaging System of the Television Set.- 11.2.1 The Cathode Ray Tube.- 11.2.2 Phosphor.- 11.2.3 The Black-and-White TV Signal.- 11.2.4 Adding Color.- 11.3 Breeder Genetic Algorithm for Optimization.- 11.3.1 Genetic Algorithm for Optimization.- 11.4 Automated Electrical Tuning of Television Sets.- 11.5 Intelligent System for Control.- 11.6 Simulation Results.- 11.7 Summary.- 12 Intelligent Manufacturing of Batteries.- 12.1 Intelligent Control of the Battery Charging Process.- 12.1.1 Problem Description.- 12.1.2 Fuzzy Method for Control.- 12.1.3 Neuro-Fuzzy Method for Control.- 12.1.4 Neuro-Fuzzy-Genetic Method for Control.- 12.2 Hardware Implementation of the Fuzzy Controller for the Charging Process.- 12.2.1 Introduction.- 12.2.2 Fuzzy Control.- 12.2.3 Implementation of the Fuzzy Controller.- 12.2.4 Experimental Results.- 12.3 Automated Quality Control of Batteries.- 12.3.1 Introduction.- 12.3.2 Fuzzy Controller.- 12.3.3 Fuzzy Control Implementation.- 12.4 Summary.