基于MIC-EEMD-改进Informer的含高比例清洁能源与储能的电力市场短期电价多步预测  

Short-term Multi-step Price Prediction for the Electricity Market With a High Proportion of Clean Energy and Energy Storage Based on MIC-EEMD-improved Informer

作　　者：许越 李强 崔晖 XU Yue;LI Qiang;CUI Hui(Key Laboratory of Research and System Evaluation of Power Dispatching Automation Technology(China Electric Power Research Institute),Haidian District,Beijing 100192,China;National Key Laboratory of Power Grid Safety(China Electric Power Research Institute),Haidian District,Beijing 100192,China)

机构地区：[1]电力调度自动化技术研究与系统评价北京市重点实验室(中国电力科学研究院有限公司),北京市海淀区100192 [2]电网安全全国重点实验室(中国电力科学研究院有限公司),北京市海淀区100192

出　　处：《电网技术》2024年第3期949-957,共9页Power System Technology

基　　金：中国电力科学研究院有限公司研究开发项目“电力现货市场通用出清与数据支撑关键技术”(52420022000R)。

摘　　要：随着电力现货市场的开展,短期电价预测对于各市场主体的决策有着重要意义,而高比例清洁能源与储能的不断接入给短期电价预测带来很大挑战。提出一种基于最大信息系数法(maximum information coefficient,MIC)、集成经验模态分解(ensemble empirical mode decomposition,EEMD)和改进Informer的短期电价多步预测模型。首先,采用MIC分析出与电价相关性较高的几类因素作为模型原始输入序列;然后,将上述原始序列进行EEMD分解后得到多条本征模函数(intrinsic mode function,IMF)和一个残余项后输入改进Informer分别得到翌日24点多步预测结果,再对预测结果进行滤波;最后,将滤波后序列的预测结果叠加得到最终的预测值。以西班牙电力市场数据进行验证,实验结果证明该模型可以有效提高电力市场短期电价多步预测精度。The emergence of the electricity spot market underscores the critical role of short-term electricity price forecasting for decision-makers in various market sectors.The increasing integration of clean energy and energy storage presents substantial challenges for short-term price predictions.This paper introduces a multi-step short-term electricity price forecasting model using the Maximum Information Coefficient(MIC),Ensemble Empirical Mode Decomposition(EEMD),and an enhanced Informer approach.Initially,MIC is applied to identify factors highly correlated with electricity prices,serving as the model's primary input sequences.These original sequences undergo EEMD decomposition,resulting in multiple Intrinsic Mode Functions(IMF)and a residual component.These components are then input separately into the improved Informer model to generate multi-step forecasts for the upcoming day,up to the 24th hour.The forecasted results undergo subsequent filtering.The filtered sequence forecast results are combined to produce the final prediction.Validation with data from the Spanish electricity market confirms that this model significantly improves the accuracy of short-term multi-step electricity price forecasting.

关 键 词：高比例清洁能源 短期电价多步预测 最大信息系数 集成经验模态分解 改进Informer 

分 类 号：TM73[电气工程—电力系统及自动化]