Fuzzy Identification base on cat swarm optimization Algorithm

A new Fuzzy Identification Method for T-S model identification algorithm is proposed, based on cats swarm and least squares method. T-S model identification is divided into structural and parameter identification. In the structure identification using cats warm can effectively overcome the traditional clustering algorithms exist slow convergence and easily fall into fall into local optimal solution, and can greatly improve the clustering convergence speed and global search capability. Then using recursive least squares method to identify parts of the model parameters, forming the Fuzzy Identification Method is based on cats swarm optimization. The method is applied to a nonlinear system to identify the system, Simulation results show that the method is effective and practical.