Active distribution network has gradually become the important direction of the future smart power grid. It plays an important role in increasing capability of renewable energy accommodation, improving the power utilization level and realizing the flexible intelligent distribution network management. The active distribution network energy management system (DMSs), which is the highest decision-making center of active distribution network, uses effective controlling and scheduling of the distributed power to guarantee global optimization operation of distribution network. This paper aims to improve the economy and reliability of the active distribution network. As the wind power and photovoltaic power generation have uncertainty, it uses random simulation technique and the penalty function method, based on the chance-constrained programming to establish a energy scheduling mathematical model, which has wind turbines, photovoltaic power generation unit and the active power energy storage device. Considering various constraint conditions, the model uses the improved particle swarm algorithm to solve. In order to verify the correctness and effectiveness of the provided model, an actual system in a certain area is used as an example and the standard particle swarm algorithm is compared.