改进变分模态分解和LSSVM的用户电力负荷预测

打开文本图片集

中图分类号：TN911.23-34；TM743 文献标识码：A 文章编号：1004-373X（2025）20-0127-08

Customer electricity load forecasting based on improved VMD and LSSVM

XIE Shixuan，LIU Liqun，WU Qingfeng （FacultyofElectronic InformationEngineering，Taiyuan UniversityofScienceand Technology，Tayuan O3Oo24，China）

Abstracts：Inordertofurtherimprovetheaccuracyofshort-termloadforecasting inpowersystemsandensure thedaily normaloperationofthepowersystems，ashort-termpowerloadforecasting modelbasedon WOA-VMD-SSA-LSSVMisproposed. The whale optimization algorithm（WOA） is used to automatically optimize the core parameters （ k -value and penalty coefficient α ） of thevariational modaldecomposition（VMD），soastogetthebestdecompositionsubsequenceandreducetheinfluenceof diferenttrendinformationontheforecastingacuracy.Theoptimized VMDisused todecomposethedata.Thesparow search algorithm （SSA）isused toimprovethemodellearningparametersoftheleastsquaressupportvectormachine（LSSVM），andthe parametersoptimizationofthepenaltyparametersandkerelfunctionsareconducted，avoidingtheproblemoflowaccuracyofa singlepredictorvariable，andestablishingaforecastingmodeltoobtainmoreaccurateforecastingresults.Thedecomposeddata isinputintothemodelseparately，andthepredictionresultsofeachsubsequencearesummeduptogetthefinalprediction results.Theexperimentalresultsshow that，incomparisonwith themodelingresultsofPSO，GWO，SABOalgorithms，the proposedmodelhas higherpredictionaccuracyandshorterconsumptiontime.Toacertainextent，itprovidesascientificdecision -making basis for load management and power optimization scheduling.

Keywords：forecastingmodel analysis；whaleoptimizationalgorithm；sparrowsearchalgorithm；variationalmodal decomposition;least squares support vector machine; data preprocessing; time series forecasting

电力产业是社会生产和人们日常生活的基础保障，高精度的电力系统短期负荷预测对保证电力系统的可靠运行有重要意义。（剩余8307字）