Neuro-Control and its Applications | springerprofessional.de

Springer Professional

nach oben

1996 | Buch

Kapitel lesen Erstes Kapitel lesen

Neuro-Control and its Applications

verfasst von: Sigeru Omatu, PhD, Marzuki Khalid, PhD, Rubiyah Yusof, PhD

Verlag: Springer London

Buchreihe : Advances in Industrial Control

Enthalten in: Professional Book Archive

Einloggen, um Zugang zu erhalten

Über dieses Buch

The series Advances in Industrial Control aims to report and encourage technology transfer in control engineering. The rapid development of control technology impacts all areas of the control discipline. New theory, new controllers, actuators, sensors, new industrial processes, computer methods, new applications, new philosophies, ........... , new challenges. Much of this development work resides in industrial reports, feasibility study papers and the reports of advance collaborative projects. The series offers an opportunity for researchers to present an extended exposition of such new work in all aspects of industrial control for wider and rapid dissemination. Sigeru Omatu, Marzuki Khalid, and Rubiyah Yusof have pursued the new developments of fuzzy logic and neural networks to present a series volume on neuro-control methods. As they demonstrate in the opening pages of their book, there is an explosion of interest in this field. Publication and patent activity in these areas are ever growing according to international is timely. databases and hence, this volume The presentation of the material follows a complementary pattern. Reviews of existing control techniques are given along side an exposition of the theoretical constructions of fuzzy logic controllers, and controllers based on neural networks. This is an extremely useful methodology which yields rewards in the applications chapters. The series of applications includes one very thorough experimental sequence for the control of a hot-water bath.

Anzeige

Inhaltsverzeichnis

Frontmatter

Chapter 1. Introduction

Abstract

Intelligent control is now becoming a common tool in many engineering and industrial applications [1], [2]. It has the ability to comprehend and learn about plants, disturbances, environment, and operating conditions [3],[4]. Some examples of the factors to be learned are plant characteristics such as its static and dynamic behaviours, some characteristics of disturbances or the environment, and equipment-operating practices [1], [2]. Figures 1.1.1 and 1.1.2 show the number of papers from INSPEC (Information Service for Physics and Engineering Communities) and patents from CASSIS (Classification for Search Support Information System), respectively [5]. From these figures, we can see the trends and the relative activities of research and applications in the field of computational and artificial intelligence. It can be observed that while research in expert systems which used to be the domain tool for intelligent systems, is declining slowly, research in neural networks is progressing rather steadily.

Sigeru Omatu, Marzuki Khalid, Rubiyah Yusof

Chapter 2. Neural Networks

Abstract

Neural networks are networks of nerve cells (neurons) in the brain. The human brain has billions of individual neurons and trillions of interconnections. Neurons are continuously processing and transmitting information to one another. In 1909, Cajal [1], [2] found that the brain consists of a large number of highly connected neurons which apparently can send very simple excitatory and inhibitory messages to each other and update their excitations on the basis of these simple messages. Figure 2.1.1 shows Purkinje Cell with its dendrite stained [2]. A neuron has three major regions; the cell body (soma), the axon, and the dendrites as shown in Fig. 2.1.2 [2].

Sigeru Omatu, Marzuki Khalid, Rubiyah Yusof

Chapter 3. Traditional Control Schemes

Abstract

In this chapter we are going to discuss several other methods of control. These methods include PID (proportional-plus-integral-plus derivative) control, self-tuning control, self-tuning PID control, generalized predictive control, and also fuzzy logic control. One of the earliest controllers that were used for control were the PI and PID controllers. PI and PID controllers have been proven to be remarkably effective in regulating a wide range of processes. The use of PI and PID controllers does not require an exact process model and hence, they are effective on industrial processes whose models are considerably difficult to derive. The PI and PID controllers are based on classical control theory and much easier to understand. Field engineers and process operators are able to relate the parameter settings and control system actions.

Sigeru Omatu, Marzuki Khalid, Rubiyah Yusof

Chapter 4. Neuro-Control Techniques

Abstract

In many real-world applications, there are many nonlinearities, unmodeled dynamics, unmeasurable noise, multiloop, etc., which pose problems to engineers in trying to implement control strategies. During the past two decades development of new control strategies has been largely based on modern and classical control theories. Modern control theory such as adaptive and optimal control techniques and classical control theory have been based mainly on linearization of systems [1]–[5].

Sigeru Omatu, Marzuki Khalid, Rubiyah Yusof

Chapter 5. Neuro-Control Applications

Abstract

In this chapter we discuss several neuro-control techniques with applications to real physical processes; a water bath temperature control system, an inverted pendulum, an electric vehicle generator control system, and a multi-input multi-output furnace. For the water bath and furnace temperature control systems, the emulator and controller neuro-control scheme is applied. However, as these real processes are slow in nature, offline learning methods are used to train the neural networks at first and then on-line learning is applied using the architecture of Fig. 4.2.5 for fine-tuning their performances. In these applications comparison is made with several traditional control methods under varying complexities in the processes. As neuro-control is relatively new it is important to see how well it compares to the more established traditional control approaches.

Sigeru Omatu, Marzuki Khalid, Rubiyah Yusof

Backmatter

Titel: Neuro-Control and its Applications
verfasst von: Sigeru Omatu, PhD
Marzuki Khalid, PhD
Rubiyah Yusof, PhD
Copyright-Jahr: 1996
Verlag: Springer London
Electronic ISBN: 978-1-4471-3058-1
Print ISBN: 978-1-4471-3060-4
DOI: https://doi.org/10.1007/978-1-4471-3058-1