Genetic optimization of fuzzy membership functions

The successful application of fuzzy control largely depends on some subjectively decided parameters, such as fuzzy membership functions. In this paper, genetic learning and turning based on real-coded genetic algorithm is proposed to automatically design and optimize the fuzzy membership function's parameters. An advantage framework, which can achieve a trade-off between execution time and optimized membership function, is introduced. By using this method, the subjectivity and blindness in the process of designing the input and output membership functions are avoided. The optimized fuzzy logic controller has been compared with the traditional one and the results demonstrate that control performance of the proposed fuzzy logic control is greatly improved.