The adaptive dynamic reconfiguration multi-objective particle swarm optimization algorithm is proposed, which uses the agglomerate hierarchical clustering algorithm to divide the population into several groups, and stores the nondominated solutions with an elite set. According to the contribution and diversity, the particles of various subgroups and the entire population are reconfigured adaptively and dynamically. At the same time, a perturbed operator is introduced to disturb non-dominated solutions set stored in the elitist set, achieving dynamically adjustment to the elite set. Test functions with different characteristics are used to verify the proposed algorithm which is compared with other similar algorithms. The results show that the proposed algorithm can speed up the convergence speed and increase the evolving ability of the population.