Abstract: An ant colony dynamic programming algorithm is presented to determine the optimal combination of a PV-BS microgrid,in order to reduce the comprehensive power generation cost of grid-connected PV-BS microgrid and the performance of optimal sizing. As the traditional heuristic algorithm is high possibility of being trapped in local optimum,the proposed to integrate ant colony algorithm and dynamic programming algorithm together,which is applied to the allocation table construction of dynamic programming to simplify the calculation process. Then,the decay parameter related to the number of iterations is introduced into the adjustment factor to improve its global search ability. Thirdly,the Boltzmann selection mechanism is introduced into the ant colony search process,and the deflection angle factor and the inflection point parameter are used to correct the ant colony pheromone update process,thereby effectively improving the optimization performance of the algorithm. Finally,the practicability and superiority of the algorithm are verified by the simulation analysis of the actual data of a residential area in Haining.