The rapid development of AM（Active Management） technology provides possibility for large-scale distributed PV（PhotoVoltaic） power generation accessing to distribution network. An optimal allocation model with the combination of AM and optimal distributed PV allocation

which takes the installation position

capacity

and AM measures as its decision variables and the maximum distributed PV energy permeability and the minimum voltage deviation as its objectives. The multi-scenario analysis based on dichotomous K-means clustering is used to deal with the uncertainties and temporal characteristics of PV output and load

which overcomes the shortage that K-means clustering scenario reduction is sensitive to the selection of initial centroid. The multi-scenario analysis and multiobjective differential evolution algorithm based on parallel computing are proposed to solve the optimization model

and the optimal PV allocation scheme and AM strategy of distribution network are obtained. The simulative results of IEEE 33-bus distribution network show that the proposed method can effectively improve the distributed PV penetration amount and ensure the power supply quality.