This paper addresses the issue of scientifically formulating an uncertainty planning strategy of the distribution network. It considers the uncertainty and relevance of distributed power sources and multiple loads. To this end a dynamic robust planning method for active distribution networks is proposed. Cholesky decomposition is used to transform the correlated random variables into mutually independent variables. Using the method of polyhedral uncertainty set representation, the uncertainty of random variables is expressed by an undistributed bounded interval. A two-stage robust planning model for an active distribution network is proposed, using the modified IEEE33-node system as an example, and the model is compared with the traditional robust planning method and without considering the correlation. The simulation results show that the method in this paper improves the rationality and economy of the uncertainty planning strategy of the distribution network. This work is supported by the Science and Technology Project of the Headquarters of State Grid Corporation of China (No. 5442PD200001).