现代信息科技2025，Vol.9Issue(21)：44-51,58,9.DOI:10.19850/j.cnki.2096-4706.2025.21.009
基于改进深度强化学习算法的风电场功率优化控制
Power Optimization Control of Wind Farms Based on Improved Deep Reinforcement Learning Algorithm
彭剑 1刘东文2
作者信息
- 1. 湖南中医药高等专科学校,湖南 株洲 412012
- 2. 广东南丰电气自动化有限公司,广东 梅州 514523
- 折叠
摘要
Abstract
To address the problem of the decrease in overall power generation in wind farms caused by wake effects and the difficulty in accurately establishing physical models of wind farms,a DRL control method that simultaneously considers wake steering control and axial induction control is proposed.This method uses the yaw angle and thrust coefficient of wind turbines as variables to implement coordinated control of the wake,mitigating the impact of wind turbine wakes and ensuring the maximization of overall power generation in wind farms.To improve the efficiency of training sample utilization,an improved DRL algorithm is proposed in this paper,which adopts a prioritized experience replay strategy on the unimproved DRL algorithm to assign different sampling priorities according to differences in the importance of experiences.Simulation on the WFSim wind farm platform demonstrates that compared with other methods,the proposed control strategy can significantly increase the active power output of wind farms.Compared with the unimproved DRL algorithm,the proposed improved DRL algorithm can avoid falling into local optima and enhance training effectiveness.关键词
风电场/尾流效应/强化学习/偏航控制/感应控制Key words
wind farm/wake effect/Reinforcement Learning/yaw control/induction control分类
信息技术与安全科学引用本文复制引用
彭剑,刘东文..基于改进深度强化学习算法的风电场功率优化控制[J].现代信息科技,2025,9(21):44-51,58,9.
