Coordinated optimization With Uncertainty Based on Model Predictive Control in Active Distribution Systems

With increasing renewable energy resources (RES) integrated, conventional distribution network (DN) is evolving to the active distribution network (ADN) with lots of controllable sources. To coordinate and optimize the deployment of these sources, many optimal models have been designed in active distribution system based on day-ahead profiles. However, most of these models have deficiency in meeting the requirements of both security and economy from ADN's operation due to the volatility of load demands and RES. Thus, a coordinated optimization model with uncertainty based on model predictive control (MPC) with robust strategy is proposed in this paper to cope with this problem. This model mainly consists of three parts: 1) robust strategy; 2) intra-rolling scheduling part; 3) real-time feedback correction part. In the first part, a day-ahead robust approach is employed to obtain control policy of the slow-motion equipment in extreme cases, which can guarantee the system security and economy. In the second part, the intra-rolling scheduling is applied to minimize the operation costs of distribution network. In the third part, real-time feedback correction model is employed to minimize the output adjustment of adjustable resources at the current time. The effectiveness and feasibility of the proposed model is demonstrated by the results based on IEEE 33.