A Tuning Method of Fuzzy Rules by Fuzzy Singleton-Type Reasoning Method. Part Ⅱ

摘要

By means of fuzzy singleton-type reasoning method, we proposed a self-tuning method for fuzzy rule generation. In this proposed approach, first we give a neuro-fuzzy learning algorithm for tuning fuzzy rules under fuzzy singleton-type reasoning method, then we roughly design initial tuning parameters of fuzzy rules based on fuzzy clustering algorithms, before learning a fuzzy model. Therefore, the learning time can be expected to be reduced and the fuzzy rules generated by the proposed approach are reasonable and suitable for the identified system model. Finally, the efficiency of the proposed method is also shown by identifying a nonlinear function.