Open AccessDeep Reinforcement Learning based Applications in Smart Power Systems
Author(s) -
Yong Chen,
Xuya Peng,
Xin Xu,
Hao Wu
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1881/2/022051
Subject(s) - reinforcement learning , computer science , artificial intelligence , electric power system , dimension (graph theory) , control (management) , machine learning , industrial engineering , power (physics) , engineering , mathematics , physics , quantum mechanics , pure mathematics
Due to the complexity, uncertainty and the increase of data dimension in power systems, conventional methods often encounter bottlenecks when trying to solve decision-making and control problems. Therefore, data-driven methods to solve these problems are being widely studied. Deep reinforcement learning (DRL) is one of these data-driven methods and is regarded as true artificial intelligence (AI). It has been applied to solve a series of complex sequential decision-making problems, including those in the power system. This paper first reviews the basic idea, model, algorithm and technology of DRL. Then its application in power systems is considered, such as energy management, demand response, electricity market, and operation control. Also, the latest development of DRL in the power system is discussed.