风光氢耦合系统的经济随机模型预测控制方法研究  

RESEARCH ON ECONOMIC STOCHASTIC MODEL PREDICTION CONTROL METHOD OF WIND-SOLAR-HYDROGEN COUPLING SYSTEM

作　　者：黄宇[1] 周新鹏 郭浩德 王嘉俊 张启亮 Huang Yu;Zhou Xinpeng;Guo Haode;Wang Jiajun;Zhang Qiliang(Department of Automation,North China Electric Power University,Baoding 071003,China)

机构地区：[1]华北电力大学自动化系,保定071003

出　　处：《太阳能学报》2024年第8期144-153,共10页Acta Energiae Solaris Sinica

基　　金：国家自然科学基金(62203172);中央高校基本科研业务费专项(2021MS089)。

摘　　要：针对风光氢耦合系统源荷两侧不确定性对系统运行造成的影响,提出一种基于经济随机模型预测控制的优化方法。根据风光氢耦合系统中设备的特性,建立计及设备启停的状态空间模型;利用场景生成技术对风光出力及电负荷预测数据进行处理,以生成描述系统不确定性的场景集;基于生成的场景集,在设计的经济随机模型预测控制框架下,构造系统的混合整数线性规划问题,进而对系统进行经济优化控制。由于经济随机模型预测控制方法需提供准确描述系统不确定性的场景集,为此提出一种基于非参数预测的场景生成机制。通过案例仿真分析,相较于常规随机模型预测控制方法,所提控制方法降低5.89%的运行成本;相较于常规模型预测控制方法,降低13.25%的运行成本,验证了所提方法能有效解决风光氢耦合系统的不确定性。An optimization method based on economic stochastic model prediction control is proposed to address the impact of uncertainty on both sides of the source and load of the wind-solar-hydrogen coupling system.Firstly,according to the characteristics of the equipment in the wind-solar-hydrogen coupling system,a state-space model considering the start-stop state of the equipment is established.Then,the scenario generation technology is used to process wind and solar output,as well as electrical load prediction data to generate a scenario set that describes the system uncertainty.Finally,based on the generated scenarios,a mixed-integer linear programming problem is formulated under the designed economic stochastic model predictive control framework,and then the system is economically optimized and controlled.A scenario generation mechanism based on nonparametric prediction is proposed,which provides a scenario set that accurately describes the system′s uncertainty for the economic stochastic model prediction control method.Simulation results demonstrate the effectiveness of the proposed method in addressing uncertainty in the wind-solar-hydrogen coupling system,achieving a 5.89%reduction in operating costs compared to conventional stochastic model predictive control method,and a 13.25%reduction compared to conventional model predictive control method.

关 键 词：风电 光伏 氢能 非参数预测 不确定性 随机模型预测控制 场景生成 

分 类 号：TM732[电气工程—电力系统及自动化]