深度强化学习增强的电力系统研究

Power Systems Enhanced by Deep Reinforcement Learning

由于电力系统的复杂性、不确定性和数据维度的增加,传统方法在解决决策和控制问题时会遇到瓶颈。因此,解决这些问题的数据驱动方法正在被广泛研究。深度强化学习是数据驱动的方法之一,被认为是真正的人工智能。深度强化学习结合了深度学习的感知能力和强化学习的决策能力。这一研究领域已被广泛应用于解决复杂的顺序决策问题,包括电力系统中的问题。本文首先介绍了深度强化学习的基本思想、模型、算法以及技术。然后介绍了在电力系统中的应用,此外,还讨论了深度强化学习在电力系统中的应用前景和挑战。

Due to the complexity and uncertainty of power system and the increase of data dimension,traditional methods encounters bottlenecks in solving decision-making and control problems.Therefore,data-driven approaches to solving these problems are being studied extensively.Deep reinforcement learning is one of the data-driven approaches that are considered to be true artificial intelligence.Deep reinforcement learning combines the perceptual ability of deep learning with the decision-making ability of reinforcement learning.This research area has been widely used to solve complex sequential decision problems,including problems in power systems.This paper first introduces the basic ideas,models,algorithms and techniques of deep reinforcement learning.Then,the application of deep reinforcement learning in power system is introduced.In addition,the application prospect and challenge of deep reinforcement learning in power system are discussed.