Modeling a fuzzy system by the integrated virtual and genetic algorithms

Modeling a fuzzy system by the integrated method of fuzzy c-means, virtual fuzzy sets, and genetic algorithms is investigated in this paper. The fuzzy c-means method is exploited to cluster the training data. Based on the clustering result, the virtual fuzzy sets can be simply constructed. The fuzzy rule base is then formed with the help of the established virtual fuzzy sets. Since the inferred results from the fuzzy model may not coincide with the desired outputs, genetic algorithms are used to optimize the membership functions. How the proposed algorithms work is discussed in detail. Simulation results show that the presented model outperforms the conventional approaches.