| 引用本文: | 王翰林,赵建立,张沛超,孙烨.光储氢集成系统的随机模型预测控制方法[J].电力自动化设备,2025,45(2):86-93,126. |
| WANG Hanlin,ZHAO Jianli,ZHANG Peichao,SUN Ye.Stochastic model predictive control of PV-storage-hydrogen integrated system[J].Electric Power Automation Equipment,2025,45(2):86-93,126. |
|
| 摘要: |
| 光储氢集成系统是提升光伏消纳率、降低电解制氢成本的重要途径之一，但电解槽复杂的非线性模型以及光伏出力的高随机性给系统运行优化带来了困难。提出电解槽的高精度凸拟合模型，该模型能计及电解槽功率和温度对效率的影响，且便于优化问题的建模与求解。将模型预测控制(MPC)与随机优化相结合，提出光储氢集成系统的二阶段随机MPC方法。建立双层求解框架，其中上层用于处理电解槽的非线性模型，下层基于交替方向乘子法实现大规模场景的解耦计算。光储氢集成系统的仿真算例结果表明，配置电池、储氢罐等储能单元显著提升了制氢的经济性和光伏消纳率；相较于常规MPC，考虑光伏和氢负荷分布信息的二阶段随机MPC进一步提升了系统经济性。 |
| 关键词: 电解槽 光储氢集成系统 最优调度 凸拟合模型 模型预测控制 随机优化 |
| DOI：10.16081/j.epae.202405029 |
| 分类号:TK912 |
| 基金项目:国家重点研发计划项目(2021YFB2401200) |
|
| Stochastic model predictive control of PV-storage-hydrogen integrated system |
|
WANG Hanlin1, ZHAO Jianli2, ZHANG Peichao1, SUN Ye1
|
|
1.Key Laboratory of Control of Power Transmission and Conversion of Ministry of Education, Shanghai Jiao Tong University, Shanghai 200240, China;2.State Grid Shanghai Municipal Electric Power Company, Shanghai 200122, China
|
| Abstract: |
| The photovoltaic(PV)-storage-hydrogen integrated system is one of the important ways to improve the PV consumption rate and reduce the cost of electrolytic hydrogen production. However, the complex nonlinear model of the electrolyzer and the high randomness of the PV power pose challenges to optimization of system operation. A high-precision convex fitting model for the electrolyzer is proposed, which considers the impact of power and temperature on efficiency and facilitates the modeling and solution of optimization problems. The model predictive control(MPC) is combined with stochastic optimization to propose a two-stage stochastic MPC method for the integrated system. Then, a two-level solution framework is established, where the upper level is used to deal with the nonlinear model of the electrolyzer and the lower level implements decoupled calculations for large-scale scenarios based on the alternating direction method of multipliers. Simulation examples of the PV-storage-hydrogen integrated system are established. The results show that configuring energy storage units such as batteries and hydrogen storage tanks significantly improves the economics of hydrogen production and the photovoltaic consumption rate. Compared with conventional MPC, the stochastic MPC considering the distribution information of photovoltaic and hydrogen loads further improves the system’s economics. |
| Key words: electrolyzer PV-storage-hydrogen integrated system optimal dispatch convex fitting model model predictive control stochastic optimization |
