Identification of a Nonlinear System by Determining of Fuzzy Rules
Subject Areas : Machine learninghojatallah hamidi 1 , Atefeh Daraei 2
1 - Department of Industrial Engineering, K. N. Toosi University of Technology, Tehran, Iran
2 - K. N.Toosi University of Technology
Keywords: Data mining, Classification, Heart disease, Diagnosis, Prognosis, Treatment,
Abstract :
In this article the hybrid optimization algorithm of differential evolution and particle swarm is introduced for designing the fuzzy rule base of a fuzzy controller. For a specific number of rules, a hybrid algorithm for optimizing all open parameters was used to reach maximum accuracy in training. The considered hybrid computational approach includes: opposition-based differential evolution algorithm and particle swarm optimization algorithm. To train a fuzzy system hich is employed for identification of a nonlinear system, the results show that the proposed hybrid algorithm approach demonstrates a better identification accuracy compared to other educational approaches in identification of the nonlinear system model. The example used in this article is the Mackey-Glass Chaotic System on which the proposed method is finally applied.