Abstract: It is a challenge to the stable operation of the Active Distribution Network with the distributed generations (DG), and the reconfiguration is an important method to improve the stability of the system. The Improved Niche Multi-objective Particle Swarm Optimization (INMPSO) was proposed to solve the model of active distribution network reconfiguration with DG; the network loss, voltage quality index and the number of switching operation were set up as the multi-objective optimization functions. Through the niche sharing mechanism which maintains the diversity and distribution uniformity of the population to update the global optimal position, the optimal compromise solution was chosen from the Pareto solutions using the fuzzy satisfaction evaluation decision method, which provides a scientific basis for the decision maker. Finally, the optimization results of IEEE33 node distribution system based on MPSO were compared to verify the effectiveness of the proposed model and method.