Dynamic economic dispatch of power system based on DDPG algorithm

• Published in: Energy reports Vol. 8; pp. 1122 - 1129
• Main Authors: Liu, Zhicheng, Liu, Yipeng, Xu, Hao, Liao, Siyang, Zhu, Kefan, Jiang, Xinxiong
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.08.2022; Elsevier
• Subjects: Artificial intelligence; DDPG; Power system dispatch; Renewable energy; DDPG; Power system dispatch; Renewable energy; Artificial intelligence
• ISSN: 2352-4847; 2352-4847
• DOI: 10.1016/j.egyr.2022.02.231

With the rapid developments of renewable energy sources, the uncertainty of the power supply will bring new challenges to scheduling problems, and conventional scheduling strategies may lose their effectiveness. The current general scheduling strategies need to model the uncertainty of the environment, while it is difficult to achieve a high degree of accuracy in the power system with high penetration rate of new energy, which will directly affect the scheduling result. In response to this problem, this paper studies the economic dispatch of power systems based on deep deterministic policy gradient (DDPG),which avoids the uncertainty modeling of the environment in principle. Combined with the basic economic dispatch model, this paper has defined a learning mode of the algorithm and built an algorithm framework of economic dispatch of power system based on DDPG. The results of the experiment show that proposed algorithm is highly adaptable to random fluctuation of renewable energy.