Particle swarm optimization algorithm(PSO, in short) is a heuristic global optimization algorithm based on swarm intelligence. Each particle of the swarm represents one candidate solution of the optimization problem. PSO searches the optimal region of optimization space through the interaction of particles. In this article, the PSO which has slow convergence rate and is easily trapped in local optimum region is modified by changing the velocity updating formula of PSO, adding the disturbance term, adding crossover and mutation operator to the algorithm so that the performance of the hybrid PSO is significantly improved. Some experimental results indicate that the improved PSO algorithm is effective and has good capability on both global and local optimization problems.