The massive access of renewable energy and electric vehicles in the active distribution network (ADN) has brought many challenges to the planning, construction, dispatch and operation of the current distribution network. By comprehensively considering the uncertain factor like source, network and load side, the research and design of the ADN optimization dispatching method have important guiding significance for the actual operation of the distribution network. Firstly, an ADN comprehensive evaluation system is established to measure the optimal dispatch level of ADN from three aspects which are active controllability, active management and active economy. Then, based on the comprehensive evaluation system, a multi-stage optimization method considering the intelligent dispatching system of electric vehicles is proposed. This method only smoothes the fluctuations caused by intermittent power supply and electric vehicles connected to the grid, but also realizes the optimal dispatch of the current distribution network structure and resources. Therefore, the proposed method can be widely used in the day-ahead scheduling of power system to realize the efficient operation and management of ADN.