Particle swarm optimization (PSO) is a global stochastic evolutionary algorithm. It tries to find optimal regions of complex searching space through the interaction of particles in the population. Based on the PSO algorithm characteristics of searching the parameter space concurrently and efficiently, the structure parameters of adaptive network-based fuzzy inference system (ANFIS) were tuned by the hybrid algorithm which combined particle swarm optimization with least-square method. The new algorithm greatly raises the convergence speed of network parameters identification and computation. It is demonstrated by numerical simulations that the designed algorithm is effective.

ANFIS,    , Particle swarm optimization,   ,  Least-square method