Learning rule structures of fuzzy controllers based on genetic algorithm.

• Autorzy: Xiong N.
• Języki publikacji: EN

Abstrakty

EN

One of the major difficulties in fuzzy control of complex processes is the "curse of dimensionality". For the sake of a reduced size of the knowledge base some rules with incomplete premise structures covering larger regions of input domain are often desirable. The paper presents a genetic algorithm based approach to searching for suitable antecedents under which specific fuzzy actions could be derived. The rule premises are coded in a flexible way allowing the presence as well as absence of an input variable in them, in combination with a certain class of input and output fuzzy sets. On the other hand, a consistency index is introduced to give a numerical evaluation of the coherence among individual rules. This index is incorporated into the fitness function of the genetic algorithm to search for a set of optimal rule premises yielding not only good control performances but also little conflict in the rule base. The effectiveness of our work is demonstrated through experiment results on an inverted pendulum.

Słowa kluczowe

EN

rule structures; complex processes; knowledge base; incomplete premise structures; consistency index; fitness function; inverted pendulum

• Wydawca: Oficyna Wydawnicza Politechniki Wrocławskiej
• Czasopismo: Systems Science
• Rocznik: 2000
• Tom: Vol. 26, nr 2
• Strony: 69--84
• Opis fizyczny: Bibliogr. 9 poz., rys.10,

Twórcy

autor

Xiong N.

• Inst. of Process Autom., Kaiserslautern Univ., Germany.