首页|基于深度确定性策略梯度算法的新型有源配电网分区协同调控方法

基于深度确定性策略梯度算法的新型有源配电网分区协同调控方法

扫码查看
随着分布式新能源在配电网中的不断普及和应用,传统的配电网调控方法已经难以满足日益复杂的新型有源配电网调控需求.为了解决这一问题,文章提出了一种基于深度学习技术的新型调控方法.为了解决新型有源配电网内新能源波动引起的局部区域有功平衡问题,设计了新型有源配电网分区方法;建立了新型有源配电网多层优化调控马尔可夫决策模型;最后,利用深度确定性策略梯度算法对优化模型进行求解.仿真结果表明,所提方法能够有效响应新型有源配电网的调控需求,提升新型有源配电网调控水平.
Novel Method for Coordinated Control of Active Distribution Network Parti-tioning Based on Deep Deterministic Policy Gradient Algorithm
With the continuous proliferation and application of distributed new energy sources in the power distribution network,traditional distribution network control methods are no longer able to meet the increasingly complex demands of active distribution network control.To ad-dress this issue,the article proposes a novel control method based on deep learning technology.Firstly,a new active distribution network zoning method is designed to solve the problem of lo-cal active power balance caused by fluctuations of new energy within the active distribution net-work.Subsequently,a multi-layer optimization control Markov decision model for the new ac-tive distribution network is established.Finally,the optimization model is solved using the deep deterministic policy gradient algorithm.Simulation results show that the proposed method can effectively respond to the control demands of the new active distribution network and enhance its control level.

New active distribution networkdynamic zoningartificial intelligencecollabora-tive controldistributed energy resources

何剑军

展开 >

中国南方电网电力调度控制中心,广东广州 510000

新型有源配电网 动态分区 人工智能 协同调控 分布式电源

2024

长江信息通信
湖北通信服务公司

长江信息通信

影响因子:0.338
ISSN:2096-9759
年,卷(期):2024.37(4)
  • 10