基于VMD-IKmeans-QLSTM的相似日光伏发电功率预测

打开文本图片集

中图分类号：TP39；TP183 文献标识码：A 文章编号：2096-4706（2025）24-0120-10

Similar-day Photovoltaic Power Prediction Based on VMD-lKmeans-QLSTM

QIN Hansen， GUO Huan （SchoolofArtificial Intelligence，JianghanUniversity，Wuhan 43oo56，China）

Abstract：Aiming at the problems of insuficient extreme weather samplesand limited model expression ability in PV power prediction，a similar-day predictionmethodbasedon VMD-IKmeans-QLSTM is proposed.Firstly，a K-meansclustering strategy with a multi-dimensional feature system is designed to ensure asuffcient numberof extreme weather samples.Then，a multi-dimensionalweightedfeaturematrixisonstructedbasdonthePearsoncoelationcoefenttoalizeaccratesilar dayselection.Themulti-scaledecompositionof powersignals sperformedusing VariationalModeDecomposition（VMD），nd theQuantumLongandShort-TermMemorynetwork（QLSTM）withquantumbit-variationalquantumcircuitsisdesignedto utilizethequantumsuperpositionstate toenhancethenonlinear modelingcapabilityforprediction.Theapplicationresultsofthis combined model in a PV power station in Xinjiang show that the model improves R2 by 41.85% 8.06% and 48.46% in cloudy， sunny and rainy/snowy conditions，respectively， compared with the conventional model.

Keywords： photovoltaic power prediction;clustering;similar-dayselection;VariationalModalDecomposition;QLSTM

0 引言

在全球气候变化和环境保护压力日益加剧的背景下，发展清洁可再生能源已成为世界各国的战略共识[1]。（剩余13491字）