基于VMD-DBO-KELM的短期电力负荷预测方法

打开文本图片集

中图分类号：TM732 文献标志码：A 文章编号：2095-2945（2025）21-0048-04

Abstract：Focusingonshort-term powerloadforecasting，thispaperproposesaninnovativemethodthatintegratesVariational ModeDecomposition（VMD），DungBeetlesOptimization （DBO）algorithmand KernelExtremeLearning Machine （KELM）.Thebasic principlesofeachcomponentalgorithmaredescribedindetail，includingVMD'sadaptivedecompositionofpowerloadsignals， DBO'sparameteroptimizationmechanismbasedondung betlebehavior，andKELM'snonlinearprocesingcapablitieswiththe helpofkernelfunctions.Afterexperimentalverification，theresultsshowthatcomparedwithtraditionalmethods，theproposed VMD-DBO-KELM modelcanachieve higherpredictionaccuracyintheshort-termpower load forecasting processshowing significantadvantage，andverifingitsefectivenessandsuperiorityTislooksforwardtothefuturedevelopmentoftisethodin thecontextofinteligentpower systems，aiming toprovidestrong technical supprtforthestableoperationof the powersystem.

Keywords：VMD;DungBeetleOptimization;KernelExtremeLearningMachine;powerloadprediction;powersystem

在当今社会，电力系统作为现代社会的基础设施，其稳定运行对于经济发展与社会生活的正常运转起着举足轻重的作用。（剩余5867字）