现代电子技术2025，Vol.48Issue(6)：168-174,7.DOI:10.16652/j.issn.1004-373x.2025.06.025
基于VMD-IDBO-LSTM的光伏功率预测模型
Photovoltaic power prediction model based on VMD-IDBO-LSTM
摘要
Abstract
In allusion to the problem of strong fluctuation and low prediction accuracy of photovoltaic power generation,a photovoltaic power prediction model based on variational mode decomposition(VMD)and improved dung beetle optimizer(IDBO)to optimize long short-term memory network(LSTM)is proposed.VMD is used to decompose the photovoltaic power time series data to obtain sub-sequences with different frequencies but certain rules,so as to reduce the fluctuation of photovoltaic power.The variable spiral search strategy,Levy flight strategy and adaptive t-distribution mutation strategy are used to improve the dung beetle optimizer.The IDBO is compared with other optimization algorithms for the performance testing.The IDBO is used to optimize the number of hidden layers in the network and the initial learning rate in LSTM to build a prediction model.The predicted values of each subsequence are added to obtain the final predicted power results.The actual examples show that in amparison with the LSTM prediction model,the DBO-LSTM prediction model,and the VMD-DBO-LSTM prediction model,VMD-IDBO-LSTM model has higher prediction accuracy and greater precision.关键词
光伏发电/功率预测/变分模态分解/改进蜣螂算法/长短期记忆网络/优化算法Key words
photovoltaic power generation/power prediction/variational mode decomposition/improved dung beetle optimizer/long short-term memory network/optimization algorithm分类
信息技术与安全科学引用本文复制引用
乔雅宁,贾宇琛,高立艾,温鹏..基于VMD-IDBO-LSTM的光伏功率预测模型[J].现代电子技术,2025,48(6):168-174,7.基金项目
河北省社科基金项目(HB22GL026) （HB22GL026）
保定市科技计划项目(2372P001,2372P002) （2372P001,2372P002）
