Multi-objective fuzzy modeling using NSGA-II

An approach to construct multiple Pareto-optimal fuzzy systems based on NSGA-II is proposed in this paper. First, in order to obtain a good initial fuzzy system, a modified fuzzy clustering algorithm is used to identify the antecedents of fuzzy system, while the consequents are designed separately to reduce computational burden. Second, a Pareto multi-objective genetic algorithm based on NSGA-II and the interpretability-driven simplification techniques are used to evolve the initial fuzzy system iteratively with three objectives: the precision performance, the number of fuzzy rules and the number of fuzzy sets. Resultantly, multiple Pareto- optimal fuzzy systems are obtained. The proposed approach is applied to two benchmark problems, and the results show its validity.