Symbiosis of fuzzy systems and genetic algorithms for modelling and control

Pao Yue-kong Library Electronic Theses Database

Symbiosis of fuzzy systems and genetic algorithms for modelling and control

 

Author: Chan, Ping-tong
Title: Symbiosis of fuzzy systems and genetic algorithms for modelling and control
Degree: Ph.D.
Year: 2000
Subject: Intelligent control systems.
Fuzzy systems.
Fuzzy logic.
Genetic algorithms.
Hong Kong Polytechnic University -- Dissertations.
Department: Dept. of Electrical Engineering
Pages: [15, 170, 31 leaves] : ill. ; 30 cm.
Language: English
InnoPac Record: http://library.polyu.edu.hk/record=b1498138
URI: http://theses.lib.polyu.edu.hk/handle/200/1418
Abstract: In recent years, the integration of Fuzzy Systems (FS) and Genetic Algorithms (GA) has gained a lot of attention and many researchers have advocated exploiting the possibilities of this union. Essentially, FS is to provide GA a platform for the incorporation of linguistic IF-THEN rules and logic reasoning, and GA's role is providing FS with non-linear optimisation. The objectives of the studies undertaken in this thesis are to validate these arguments and to propose modelling and control algorithms for some classes of non-linear systems. The study will focus on problems such as optimisation speed, stability issues before and after optimisation, and high dimension systems. The findings are verified through benchmark simulation and experimental trials. The studies are carried out in three phases. In the first phase, a method to model a plant's data is proposed. An Antecedent Validity Adaptation (AVA) principle is introduced to refine Table Look-up (TL) scheme. The term Table Look-up with Validity (TLV) is coined as a scheme for modelling recorded data. The proposed technique is endowed with the ability of using the dormant and hidden information in the available data. An on-line algorithm is also presented. The design of an Optimised Fuzzy Logic Controller (OFLC) is introduced next. With the aid of genetic algorithms, optimised rules of fuzzy logic controller are designed. The use of symmetric properties to save the number of learning rules is described. The proposed algorithm has been applied to both linear and non-linear systems. This algorithm cuts the vast tangle of the rule-table down to size by incorporating the physical properties of systems and mechanisms of GA. Then OFLC is employed to control a beam-and-ball balance hardware test-bed. A Kalman filer controller (KFC) and a manually tuned fuzzy logic controller (MFLC) are also developed to assess the effectiveness of the OFLC. The proposed controller provides favourite performance in a systematic design manner. In the second phase, a supervisory control technique is applied to an integrated Sliding Mode Control (SMC)/FS. An Indirect Adaptive Fuzzy Sliding Mode Control (IAFSMC) with supervisory controller is proposed which guarantees the boundedness of states of the controlled system and parameter projection to avoid the divergence of the parameters. Along this line, an integrated stable FS+GA system is proposed, in which the FS is optimised by GA. A state supervisory controller, studied in the IAFSMC, is used to guarantee the stability. In the third phase, the use of Hierarchical and Fused FS (HFFS) will be discussed to alleviate the explosion of dimension. The design of an Optimised Hierarchical and Fused Fuzzy System (OHFFS) will be introduced which uses the divide-and-conquer technique to separate a large GA/FS optimisation problem into a series of more manageable problems. The fused FS is applied to the cart pole system and the hierarchical FS is employed to the truck-and-trailer system.

Files in this item

Files Size Format
b14981385.pdf 8.028Mb PDF
Copyright Undertaking
As a bona fide Library user, I declare that:
  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

     

Quick Search

Browse

More Information