The standard particle swarm optimization (PSO) algorithm is a novel evolutionary algorithm in which each particle studies its own previous best solution and the groups previous best solutions to optimization problems. One problem in PSO is its tendency of trapping into local optima. In this paper, a multi-swarm technique based on fast particle swarm optimization (FPSO) algorithm is proposed by introducing crossover operation. FPSO is global search algorithm which can prevent PSO from trapping into local optima in light of Cauchy mutation. Though it can get high optimizing precision, the convergence rate is not satisfactory. FMSO can not only find satisfying solutions, but also speed up the search.