This paper proposes a PSO-based multi-objective optimization named as DCMOPSO (dynamic changing multi-objection particle swarm optimization). In this scheme, the inertia weight and acceleration coefficients dynamic changing to explore the search space more efficiently. The crowding distance and mutation operator mechanism also adopted to maintain the diversity of nondominated solutions. The performance of DCMOPSO is investigated by some benchmark functions and compared with MOPSO and NSGA. The results indicate that DCMOPSO is feasible and competitive to get better distribute nondominated solutions.