基于VMD-SSA-BiLSTM网络下的短期电力负荷预测  

Short-term power load forecasting based on VMD-SSA-BiLSTM network

作　　者：王斌斌 孙丽江(指导)[1] WANG Binbin;SUN Lijiang(School of Business,Shanghai Dianji University,Shanghai 201306,China)

机构地区：[1]上海电机学院商学院,上海201306

出　　处：《上海电机学院学报》2024年第5期274-279,298,共7页Journal of Shanghai Dianji University

摘　　要：短期电力负荷预测是电力系统运控的重要部分,为提高负荷预测精度,针对实际负荷数据非线性、随机性等特征,建立了一种基于变分模态分解(VMD)下麻雀搜索算法(SSA)优化的双向长短期记忆网络(BiLSTM)的短期电力负荷预测模型。采用VMD对电力负荷数据进行分解,提取多个不同频率特征的模态分量,并引入SSA算法对BiLSTM网络参数进行优化,根据输入的模态分量建立SSA-BiLSTM预测模型进行预测。结果表明:相比于BiLSTM模型和VMD-BiLSTM模型,所建立的模型预测精度更高,拟合效果更好。Short-term power load forecasting is an important part of power system operation and control.To improve the accuracy of load forecasting and address the nonlinearity and randomness of actual load data,a short-term power load forecasting model is proposed based on the variational modal decomposition(VMD)and a bidirectional long short-term memory network(BiLSTM)optimized by the sparrow search algorithm(SSA).First,the VMD is used to decompose the power load data to extract multiple modal components with different frequency characteristics.Second,the SSA algorithm is introduced to optimize the BiLSTM network parameters.Then,the SSA-BiLSTM prediction model is established according to the input modal components for prediction.The results show that the proposed model has higher prediction accuracy and better fitting performance than the BiLSTM model and the VMDBiLSTM model.

关 键 词：短期电力负荷预测 变分模态分解 麻雀搜索算法 双向长短期记忆网络 

分 类 号：TM715[电气工程—电力系统及自动化]