A new hybrid algorithm GSAPSO based on the combination of particle swarm optimization(PSO)and gravitational search algorithm(GSA) was proposed for T-S model identification.The main idea is to integrate the ability of exploration of PSO with the ability of exploitation of GSA to synthesize the advantage of both algorithms.The PSO algorithm was introduced into GSA and an improved weight algorithm was presented.The structure identification and parameter identification of T-S model was realized together by using the new hybird algorithm GSAPSO and the clustering method.The results show the hybrid algorithm GSAPSO is of better capability of global optimization and higher precision than the standard PSO and GSA.