Abstract:

As the gravitational search algorithm plays a negative influence on the search accuracy of the complex issues,especially the poor search quality of standard Gravitational Search Algorithm(GSA) in the high dimensional function optimization,it is easy to get into premature convergence in the optimization process.An improved gravitational search algorithm based on genetic mutations(gmGSA) is proposed to identify the parameter of T-S model.An improved fuzzy c-means(FCM) based on simulated annealing(SA) and genetic algorithm(GA),denote as SAGA-FCM,is also proposed to identify the structure of T-S model.The simulation results show the proposed methods can effectively obtain compact and accurate fuzzy models with excellent capability of generalization.

Key words: identification of T S model;genetic mutations;FCM clustering;gravitational search algorithm