| 引用本文: | 周雪松,张宇轩,马幼捷,等.融入SAC算法的光储微网混合储能自驱优级联自抗扰控制[J].电力系统保护与控制,2025,53(13):93-104.[点击复制] |
| ZHOU Xuesong,ZHANG Yuxuan,MA Youjie,et al.Self-driven optimal cascade active disturbance rejection control for PV-storage microgrid with hybrid energy storage integrated with the SAC algorithm[J].Power System Protection and Control,2025,53(13):93-104[点击复制] |
|
| 摘要: |
| 母线电压稳定是实现新能源高水平消纳的重要前提。针对光储直流微电网混合储能系统因源荷不确定性扰动导致的母线电压波动问题，提出一种融入深度强化学习柔性动作评价(soft actor-critic, SAC)算法的自驱优级联自抗扰控制策略。首先，设计了级联扩张状态观测器来实时估计和补偿系统中的不确定性扰动，以提升系统的扰动估计精度。其次，针对系统建立了马尔可夫决策模型，并设计了状态奖励与信息熵综合评估的SAC智能体，融入控制器参数优化中。通过其在线学习和经验回放实现了控制参数的自驱优整定，进一步提高了系统的抗扰性和鲁棒性。最后，通过仿真实验对比了3种控制策略在典型工况下的控制性能，验证了所提策略的有效性和优越性。 |
| 关键词: 光储直流微电网 混合储能 自抗扰控制 深度强化学习 SAC算法 |
| DOI：10.19783/j.cnki.pspc.241244 |
| 投稿时间：2024-09-13修订日期：2025-01-13 |
| 基金项目:国家自然科学基金重大项目资助(U24B6011)；
国家自然科学基金重点项目资助(U23B20142) |
|
| Self-driven optimal cascade active disturbance rejection control for PV-storage microgrid with hybrid energy storage integrated with the SAC algorithm |
| ZHOU Xuesong1,ZHANG Yuxuan1,MA Youjie1,WANG Xinyue1,TAO Long1,WEN Hulong2,3 |
| (1.Tianjin Key Laboratory of New Energy Power Conversion, Transmission and Intelligent Control (Tianjin University
of Technology), Tianjin 300384, China; 2. Tianjin Ruineng Electric Co., Ltd., Tianjin 300385, China;
3. Tianjin Ruiyuan Electric Co., Ltd., Tianjin 300308, China) |
| Abstract: |
| Bus voltage stability is a critical prerequisite for achieving high-level integration of new energy. To address the problem of bus voltage fluctuations in PV-storage DC microgrid with hybrid energy storage systems caused by source-load uncertainties, a self-driven optimal cascade active disturbance rejection control strategy integrated with the soft actor-critic (SAC) deep reinforcement learning algorithm is proposed. First, a cascaded extended state observer is designed to estimate and compensate for system uncertainties in real time, improving the accuracy of disturbance estimation. Then, a Markov decision model is established for the system, and a SAC agent, designed with a comprehensive evaluation of state rewards and information entropy, is integrated into the controller parameter optimization. By leveraging online learning and experience replay, the control parameters are autonomously and optimally tuned, further enhancing the system’s disturbance rejection capability and robustness. Finally, the performance of three control strategies under typical working conditions is compared by simulation experiment, validating the effectiveness and superiority of the proposed approach. |
| Key words: PV-storage DC microgrid hybrid energy storage active disturbance rejection control deep reinforcement learning SAC algorithm |
