摘要
针对新能源发电量预测中单一模型精度不足的问题,提出了一种EMD-GM-Elman(empirical mode decompositiongrey model-elman)神经网络组合模型。该模型通过经验模态分解(empirical mode decomposition,EMD)预处理数据,提取局部特征;利用灰色预测模型预测各本征模态函数(intrinsic mode functions,IMF),结果输入Elman神经网络捕捉动态特征;最终通过数据重构得出预测结果。仿真结果显示,该模型预测精度从传统模型的58.1%提高到65.14%。
Aiming to address insufficient accuracy of the single model in the prediction of new energy power generation,a combined EMD-GM-Elman(Empirical Mode Decomposition-Grey ModelElman)neural network model is proposed in this paper.The model pre-processes the data through Empirical Mode Decomposition(EMD)to extract local features,and predicts the Intrinsic Mode Functions(IMFs)using the gray prediction model,and inputs the results into the Elman neural network to capture the dynamic features,and then reconstructs the results through data reconstruction to obtain the prediction results.Finally,the prediction results are obtained through data reconstruction.The simulation results show that the prediction accuracy of the model is improved from 58.1%to 65.14%of the traditional model.
作者
赵汉超
从兰美
刘杰
韩子月
胡宁宁
潘广源
夏远洋
ZHAO Hanchao;CONG Lanmei;LIU Jie;HAN Ziyue;HU Ningning;PAN Guangyuan;XIA Yuanyang(School of Automation and Electrical Engineering,Linyi University,Linyi 276000,Shandong,China;Yalong River Basin Hydropower Development Co.,Ltd.,Chengdu 610000,Sichuan,China)
出处
《电网与清洁能源》
CSCD
北大核心
2024年第10期132-141,共10页
Power System and Clean Energy
基金
国家自然科学基金(62103177)。
