基于分布式模型预测控制的电-氢混合储能系统控制策略  

Control Strategy for Electricity-Hydrogen Hybrid Energy Storage System Based on Distributed Model Predictive Control

作　　者：余晋宇 周荔丹[1] 于天佑 王子强 赵轩辉 姚钢[1] YU Jinyu;ZHOU Lidan;YU Tianyou;WANG Ziqiang;ZHAO Xuanhui;YAO Gang(School of Electronic Information and Electrical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China)

机构地区：[1]上海交通大学电子信息与电气工程学院,上海市200240

出　　处：《电力系统自动化》2025年第3期114-124,共11页Automation of Electric Power Systems

基　　金：中核集团领创基金资助项目(22GFC-JJ12-476)。

摘　　要：储能系统具备平抑新能源波动、储存多余电能等作用,但单一类型的储能单元因响应速度、容量等限制难以满足需求。为提升孤岛直流微电网新能源消纳能力,提出了一种基于分布式模型预测控制(DMPC)的电-氢混合储能系统控制方案,可充分发挥锂电池高功率密度和一体化可再生燃料电池高能量密度的优势。首先,针对电、氢储能子系统分别设计DMPC控制器,可实现正常工况下电、氢储能分别支撑高、低频负荷,降低电储能充放电次数,延长寿命;而在氢储能停机工况下,可实现电储能单独支撑负荷,维持系统稳定。其次,考虑电、氢储能响应速度差异,基于DMPC代价函数的荷电状态恢复项,引入了权重自适应算子,避免了负责快速响应的电储能平均电压恢复能力突降,提高了直流电压暂态稳定性。最后,通过Simulink仿真验证了所提控制策略的有效性。Energy storage systems can smooth out fluctuations in renewable energy and store excess electricity.However,a single type of storage unit cannot meet demand due to limitations in response speed and capacity.To increase the capacity of islanded DC microgrids to accommodate renewable energy,this paper proposes a control scheme for an electricity-hydrogen hybrid energy storage system based on distributed model predictive control(DMPC).This scheme makes full use of the high power density of lithium batteries and the high energy density of unitized renewable hydrogen fuel cells.Firstly,DMPC controllers are designed for the electricity and hydrogen energy storage subsystems.Under normal operation conditions,the electricity and hydrogen energy storage systems respectively support high-and low-frequency loads,reducing the number of charging-discharging cycles for the batteries and extending their lifespan.Under the condition of hydrogen energy storage system shutdown,the electricity energy storage system independently supports the load to maintain system stability alone.Secondly,considering the different response speeds of the electricity and hydrogen energy storage systems,this paper introduces a weight-adaptive operator into the state-ofcharge recovery term of the DMPC cost function.It prevents a significant drop in the average voltage recovery capability of the electricity energy storage system,which is responsible for rapid response,thereby improving DC voltage transient stability.Finally,the effectiveness of the proposed control strategy is verified through simulation by Simulink.

关 键 词：孤岛直流微电网 分布式模型预测控制 电-氢混合储能 燃料电池 暂态稳定 

分 类 号：TM91[电气工程—电力电子与电力传动]