적응 확률을 갖는 유전자 알고리즘을 사용한 퍼지규칙의 최적화

Fuzzy Rule Optimization Using Genetic Algorithms with Adaptive Probability

  • 발행 : 1996.06.01

초록

퍼지제어에서 퍼지규칙은 퍼지제어기의 제어결정을 내리는데 중요한 역할을 한다. 그래서, 제어성능은 주로 퍼지규칙의 질에 의해서 결정된다. 본 논문에서 우리는 교차와 돌연번이의 확률이 적응적으로 변화되는 유전자 알고리즘을 사용하여 퍼지규칙을 최적화 하는 방법을 기술한다. 또한 본 논문에서 우리는 플랜트의 응답을 듀개의 부분으로 나누어 제어 목적을 만족하게 하는 적합도 측정 방식을 제안한다. 좀더 빠른 해답을 얻기 위해 우리는 초기의 퍼지규칙으로 무작위적인 규칙을 사용하지 않고 자동으로 퍼지규칙을 생성하는 방법을 사용하여 초기 퍼지규칙으로 사용했다. 이렇게 얻어진 퍼지규칙이 좋은 것인지를 보여주기 위해 비선형 플랜트를 이용하여 시뮬레이션 해보았다. 시뮬레이션 결과 우리의 방법이 합리적이고 유용한 것임이 밝혀졌다.

Fuzzy rules in fuzzy logic control play a major role in deciding the control dynamics of a fuzzy logic controller. Thus, control performance is mainly determined by the quality of fuzzy rules. This paper introduces an optimization method for fuzzy rules using GAS with adaptive probabilies of crossover and mutation. Also we design two fitness measures to satisfy control objectives by partitioning the response of a plant into two parts. An initial population is generated by an automatic fuzzy rule generation method instead of random selection for fast a.pproaching to the final solution. We employed a nonlinear plant to simulate our method. It is shown through simulation that our method is reasonable and can be useful for optimizing fuzzy rules.

키워드

참고문헌

  1. IEEE Trans. on Systems, Man and Cybernetics v.20 Fuzzy Logic in Control Systems : Fuzzy Logic Controller-Part Ⅰ/Ⅱ C.C.Lee
  2. Journal of Intelligent and Fuzzy Systems v.1 Defuzzification in Fuzzy Controllers H.Hellendoorn;C.Thomas
  3. Fuzzy Control and Fuzzy Systems W.Pedrycz
  4. IEEE Trans. on Neural Networks v.3 Self-Learning Fuzzy Controllers Based on Temporal Back Propagation J.S.R.Jang
  5. IEEE Trans. on Neural Networks v.3 Learning and Tuning Fuzzy Logic Controllers Through Reinforcements H.R.Verenji;P.Khedkar
  6. IEEE Trans. on Computers v.40 Neural-Network-Based Fuzzy Logic Control and Decision System C.T.Lin;C.G.Lee
  7. Proceedings of the International Conference on Fuzzy Logic Intelligent Control Based on Fuzzy Logic and Neural Net Theory C.C.Lee
  8. Journal of the Korea Fuzzy Logic and Intelligent Systems Society v.5 Automatic Fuzzy Rule Generation by Simulating Human Knowledge Gathering Process S.H.Jung
  9. Proceedings of the Korea Fuzzy Math and Systems Society v.4 Automatic Fuzzy Rule Generation by Simulating Human Control Strategies S.H.Jung;T.G.Kim;K.H.Park
  10. IEEE Computer Magazine Genetic Algorithms : A Survey M.Srinivas;L.M.Patnaik
  11. IEEE Computer Magazine Genetic-Algorithms Programming Environments J.L.R.Filho;P.C.Treleaven
  12. IEEE Trans. on Systems, Man and Cybernetics v.21 A Genetic Algorithm for the Linear Transportation Problem G.A.Vignaux;Z.Michalewicz
  13. IEEE Trans. on Fuzzy Systems v.1 Fuzzy Control of pH Using Genetic Algorithms C.L.Karr;E.J.Gentry
  14. IEEE Trans. on Systems, Man and Cybernetics v.24 Adaptive Probabilities of Crossover and Mutation in Genetic Algorithms M.Srinivas;L.M.Patnaik
  15. Electronics Letters v.30 Defuzzification Method for Multishaped Output Fuzzy Sets S.H.Jung;K.H.Cho;T.G.Kim;K.H.Park