基于需求响应的电力负荷预测与建模研究  

Research on Load Forecasting and Modeling of Demand Response Uncertainty

作　　者：宋振培 张少迪 周品品 SONG Zhenpei;ZHANG Shaodi;ZHOU Pinpin(Shanghai University of Electric Power,Shanghai 200090 China;Shanghai Electrical Appliances Research Institute(Group)Co.,Ltd.,Shanghai 200063,China;Shanghai Key Laboratory of Smart Grid Demand Response,Shanghai 200063,China)

机构地区：[1]上海电力大学,上海200090 [2]上海电器科学院研究所(集团)有限公司,上海200063 [3]上海市智能电网需求响应重点实验室,上海200063

出　　处：《上海电力大学学报》2024年第1期51-57,共7页Journal of Shanghai University of Electric Power

基　　金：上海市中央引导地方科技发展资金项目(YDZX20223100003002)。

摘　　要：电力负荷不确定性问题是需求响应不确定性问题的重要研究方向之一。准确、有效的电力系统负荷预测是保障电网安全、稳定运行和社会正常生产的重要前提。为了提高电力系统负荷预测的准确性,提出了HI-Informer模型进行电力负荷预测。该模型通过最小二乘法将历史惯性(HI)模型和Informer模型融合。实验结果显示,HI-Informer模型在电力负荷预测中表现出色,其预测结果与实际观测值高度吻合,成功捕捉了电力负荷趋势和季节性变化。相比之下,其他模型在某些情况下出现了明显的误差,未能有效捕捉相同特征。The study of power load uncertainty is one of the important research directions of demand response uncertainty.The accurate and effective forecasting of electrical system load is a crucial prerequisite for ensuring the safe and stable operation of the power grid and the normal functioning of society.To enhance the accuracy of load forecasting,this study proposes the HIInformer model for electric load forecasting,which combines historical inertia modeling with the Informer model using the least squares method.The experimental results demonstrate that the HIInformer model excels in electric load forecasting.Its predictions closely match the actual observed values and successfully capture the trends and seasonal variations in electrical load.In contrast,other models exhibit noticeable errors in certain scenarios and fail to effectively capture the same characteristics.

关 键 词：需求响应 电力负荷 长短期记忆 历史惯性 

分 类 号：TM614[电气工程—电力系统及自动化] TM712