There are problems of distribution network voltage over-limit, increased network loss and curtailment of wind and solar power caused by a distributed generation (DG) connected to the grid. Thus an active and reactive power coordinated optimization method for an active distribution network is proposed. It considers demand response (DR) and three-terminal soft open point (SOP) with energy storage system (ESS). First, considering the uncertainty of renewable energy output, day-ahead scenarios are generated based on conditional generative adversarial networks (CGAN). Then, DR and a three-terminal SOP with ESS are modeled and coordinated with the DG and traditional reactive power compensation equipment. A mathematical model is constructed with the optimization objectives of minimum network loss, minimum voltage deviation, minimum curtailment of wind and solar power, and maximum user comfort and user economy. After that, it is transformed into a mixed integer second-order cone programming model and analyzed by GUROBI. Finally, simulation is carried out on an improved IEEE 33-node and IEEE 69-node distribution system. The results show that the proposed optimization method, which takes into account the uncertainty of wind and solar power, improves the economy and reliability of distribution network operation while giving consideration to the comfort and economy of users.