Abstract: In order to improve the utilization rate of wind and light energy and reduce the impact of its output volatility on the distribution network, the active distribution system model with wind-photovoltaic-energy storage hybrid generation system is established by using the time and space complementarity between wind and solar energy and the soothing effect of energy storage device on power. With the reactive compensation device as the optimization variable, multi-period dynamic reactive optimization of the active distribution network is established, the multi-objective optimization model with the lowest network loss and voltage deviation is established, and the optimization model is converted into a second-order cone programming（SOCP） model with convex feasible domain by using the relaxation technology. Finally, the simulation is carried out on IEEE 33-node distribution system, the reactive optimization is conducted for distribution network with different types of distributed power generation modes, the role of the active distribution network pressure control and loss reduction is compared with the analysis, and the optimization strategy of wind-photovoltaic-energy storage hybrid generation system is analyzed. Finally, the accuracy and reliability of second-order cone programming relaxation technology are verified.