Decentralized Robust Optimization for Real-time Dispatch of Power System Based on Approximate Dynamic Programming

摘要

Responding to the uncertainty challenges of high proportion wind power and the privacy of multi-area power system, a decentralized robust optimization method is proposed. Firstly, the uncertainty of wind power is considered by wind power prediction interval. The uncertain information of wind power is converted into boundary information of prediction interval by robust optimization. Secondly, the automatic generation control (AGC) unit is used to deal with the fluctuation of wind power. By optimizing the participation factors of AGC units, the robust dispatch model is formulated. Then, a decentralized robust optimization method is formulated based on approximate dynamic programming (ADP). The proposed method decouples centralized problem by approximate value function. Finally, a real 4-region 2298-node power system was tested to demonstrate the effectiveness of the proposed method.