为充分挖掘新能源调频潜力,该文提出基于深度强化学习的风电机组快速频率响应优化策略。首先,基于时域解析建模方法,构建计及调频死区的风力机动态频率响应解析模型,通过特征参数解耦分析,揭示死区阈值、一次调频增益系数与风电渗透率等多维参数与电网频率偏差的关联机制,完成多参数联合作用下频率动态特性的多维关联性量化表征;其次,针对多参数耦合、工况时变条件下传统调频相参数整定难以辨识最优解集的问题,设计基于深度确定性策略梯度（DDPG）的参数动态寻优框架,通过智能体与环境的交互学习,实现参数的自适应调整。最后,基于含风力机并网的电力系统仿真模型进行验证,结果表明：在不同风速和不同扰动场景下,相较于传统参数设置与粒子群算法,所提策略具有更优的频率支撑性能与工况适应能力。
Abstract
With the increasing penetration of wind power, the decrease in system inertia raises concerns about frequency stability. To fully utilize the frequency regulation potential of renewable energy, this paper proposes an optimization strategy for rapid frequency response of wind turbines based on deep reinforcement learning. Firstly, a dynamic frequency response model of wind turbines, considering the frequency regulation dead zone, is constructed using time-domain analysis. The model reveals the correlation between multiple parameters, such as the dead zone threshold, primary frequency regulation gain, and wind power penetration, and grid frequency deviation, through decoupling analysis. By this method, establish a multi-dimensional correlation quantitative characterization of frequency dynamic characteristics under the combined action of multiple parameters. Secondly, to address the challenge of identifying the optimal solution set in traditional frequency regulation methods, a dynamic parameter optimization framework based on DDPG is designed, enabling adaptive parameter adjustment through agent-environment interaction. Finally, simulations of a power system with wind turbines connected to the grid verify that the proposed strategy performs better in frequency support and adaptability than traditional parameter settings and particle swarm algorithms under different wind speeds and disturbance scenarios.
关键词
风电机组 /
风电调频 /
快速频率支撑 /
多维参数耦合 /
深度强化学习 /
参数优化
Key words
wind turbines /
wind turbine frequency regulation /
fast frequency support /
multi-dimensional parameter coupling /
deep reinforcement learning /
parameter optimization
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 李军徽, 冯喜超, 严干贵, 等. 高风电渗透率下的电力系统调频研究综述[J]. 电力系统保护与控制, 2018, 46(2): 163-170.
Li J H, Feng X C, Yan G G, et al.Survey on frequency regulation technology in high wind penetration power system[J]. Power System Protection and Control, 2018, 46(2): 163-170.
[2] 钱敏慧, 张建胜, 秦文萍, 等. 高比例风电联网背景下风电机组快速频率支撑研究综述[J]. 太阳能学报, 2025, 46(10): 714-726.
Qian M H, Zhang J S, Qin W P, et al.Rewiew of fast frequency support of WTGs under background of high proportion wind power grid-connection[J]. Acta Energiae Solaris Sinica, 2025, 46(10): 714-726.
[3] 张正阳, 殷明慧, 李阳, 等. 考虑机组稳定约束的风机一次调频控制策略[J]. 中国电机工程学报, 2023, 43(4): 1471-1480.
Zhang Z Y, Yin M H, Li Y, et al.Turbine stability-constrained primary frequency control of wind turbine generator[J]. Proceedings of the CSEE, 2023, 43(4): 1471-1480.
[4] 李军, 杨德健, 林旭, 等. 基于转矩极限控制的双馈风机改进频率控制策略[J]. 智慧电力, 2023, 51(8): 53-58, 74.
Li J, Yang D J, Lin X, et al.Improved frequency control strategy of double-fed induction generator based on torque limit control[J]. Smart Power, 2023, 51(8): 53-58, 74.
[5] 颜湘武, 王德胜, 杨琳琳, 等. 直驱风机惯量支撑与一次调频协调控制策略[J]. 电工技术学报, 2021, 36(15): 3282-3292.
Yan X W, Wang D S, Yang L L, et al.Coordinated control strategy of inertia support and primary frequency regulation of PMSG[J]. Transactions of China Electrotechnical Society, 2021, 36(15): 3282-3292.
[6] 王凡, 李海峰, 胥国毅, 等. 调频关键参数对电网频率特性的影响及其灵敏度分析[J]. 电力系统保护与控制, 2020, 48(20): 1-8.
Wang F, Li H F, Xu G Y, et al.Influence of key parameters of frequency control on frequency characteristics of power grid and sensitivity analysis[J]. Power System Protection and Control, 2020, 48(20): 1-8.
[7] 刘洋, 邵广惠, 张弘鹏, 等. 新能源参与系统一次调频分析及参数设置[J]. 电网技术, 2020, 44(2): 683-688.
Liu Y, Shao G H, Zhang H P, et al.Analysis of renewable energy participation in primary frequency regulation and parameter setting scheme of power grid[J]. Power System Technology, 2020, 44(2): 683-688.
[8] 李东东, 陈治廷, 徐波, 等. 考虑一次调频死区的风储协同调频机理分析及死区参数优化[J]. 电力系统保护与控制, 2024, 52(4): 38-47.
Li D D, Chen Z T, Xu B, et al.Mechanism analysis and optimal dead zone parameter of wind-storage combined frequency regulation considering primary frequency regulation dead zone[J]. Power System Protection and Control, 2024, 52(4): 38-47.
[9] 曹俊英, 姚骏, 刘育明, 等. 考虑一次调频死区的风电机组虚拟同步控制参数优化研究[J]. 电工电能新技术, 2023, 42(12): 29-38.
Cao J Y, Yao J, Liu Y M, et al.Optimization of virtual synchronous control parameter for WTG considering dead zone of primary frequency regulation[J]. Advanced Technology of Electrical Engineering and Energy, 2023, 42(12): 29-38.
[10] 盛四清, 占志刚, 吴林林, 等. 考虑频率二次跌落的风电机组调频控制研究[J]. 太阳能学报, 2023, 44(8): 485-491.
Sheng S Q, Zhan Z G, Wu L L, et al.Research on frequency regulation control of wind turbines considering secondary frequency drop[J]. Acta Energiae Solaris Sinica, 2023, 44(8): 485-491.
[11] 杨强, 仲卫, 谢善益, 等. 基于粒子群算法优化PID的海上风电柔直并网系统调频控制[J]. 自动化技术与应用, 2024, 43(8): 144-148.
Yang Q, Zhong W, Xie S Y, et al.Frequency regulation control of offshore wind power flexible and direct grid-connected system based on particle swarm optimization PID[J]. Techniques of Automation and Applications, 2024, 43(8): 144-148.
[12] 孙琦, 李岩, 冀宵玉, 等. 基于改进粒子群算法的风机下垂控制研究[J]. 电气自动化, 2022, 44(5): 29-31.
Sun Q, Li Y, Ji X Y, et al.Research on droop control of wind turbine based on improved particle swarm algorithm[J]. Electrical Automation, 2022, 44(5): 29-31.
[13] 胡维昊, 曹迪, 黄琦, 等. 深度强化学习在配电网优化运行中的应用[J]. 电力系统自动化, 2023, 47(14): 174-191.
Hu W H, Cao D, Huang Q, et al.Application of deep reinforcement learning in optimal operation of distribution network[J]. Automation of Electric Power Systems, 2023, 47(14): 174-191.
[14] Kundur P.Power system stability and control[M]. New York: McGraw-Hill Professional, 2005: 271-274.
[15] 李卫国, 梁智豪. 基于风机转子转速以及荷电状态的变系数风储联合系统参与一次调频控制策略[J]. 东北电力大学学报, 2025, 45(1): 83-94.
Li W G, Liang Z H.Control strategy of the variable coefficient wind-storage combined system participating in power system primary frequency regulation based on the rotor speed and the state of charge[J]. Journal of Northeast Electric Power University, 2025, 45(1): 83-94.
[16] 杨博, 罗志彪, 胡袁炜骥. 基于人工智能的移动储能规划与调度综述[J]. 东北电力大学学报, 2025, 45(1): 1-11.
Yang B, Luo Z B, Hu Y W J. Comprehensive review on artificial intelligence based planning and scheduling for mobile energy storage systems[J]. Journal of Northeast Electric Power University, 2025, 45(1): 1-11.
[17] 孙冉, 王建波, 马彦钊, 等. 基于强化学习的新能源场站储能一次调频自适应控制策略[J]. 储能科学与技术, 2024, 13(3): 858-869.
Sun R, Wang J B, Ma Y Z, et al.Adaptive control strategy for primary frequency regulation for new energy storage stations based on reinforcement learning[J]. Energy Storage Science and Technology, 2024, 13(3): 858-869.
[18] 刘晓明, 刘俊, 姚宏伟, 等. 基于VSG的风光水火储系统频率调节深度强化学习方法[J]. 电力系统自动化, 2025, 49(9): 114-124.
Liu X M, Liu J, Yao H W, et al.Deep reinforcement learning method for frequency regulation of wind-solar-hydro-thermal-storage systems based on virtual synchronous generator[J]. Automation of Electric Power Systems, 2025, 49(9): 114-124.
[19] 范培潇, 柯松, 杨军, 等. 基于改进多智能体深度确定性策略梯度的多微网负荷频率协同控制策略[J]. 电网技术, 2022, 46(9): 3504-3514.
Fan P X, Ke S, Yang J, et al.Load frequency coordinated control strategy of multi-microgrid based on improved MA-DDPG[J]. Power System Technology, 2022, 46(9): 3504-3514.
[20] 张峰, 游欢欢, 丁磊. 新能源一次调频死区影响机理建模及系数修正策略[J]. 电力系统自动化, 2023, 47(6): 158-167.
Zhang F, You H H, Ding L.Influential mechanism modelling of dead band in primary frequency regulation of renewable energy and its coefficient correction strategy[J]. Automation of Electric Power Systems, 2023, 47(6): 158-167.
基金
吉林省自然科学基金（YDZJ202401566ZYTS）
PDF(2090 KB)
