现代电子技术2024，Vol.47Issue(11)：131-139,9.DOI:10.16652/j.issn.1004-373x.2024.11.022
基于SVMD-ISSA-CNN-TGLSTM的供热负荷预测模型
Heat load prediction model based on SVMD-ISSA-CNN-TGLSTM
摘要
Abstract
In the current research on centralized heating load prediction,the internal factors of heat exchange stations and the low accuracy of heating load prediction are rarely considered,so a hybrid prediction model based on SVMD-ISSA-CNN-TGLSTM is proposed.A CNN-TGLSTM model with spatial extraction capability is constructed by convolutional neural network(CNN)and transformation-gated long short-term memory(TGLSTM)neural network.In view of the non-stationary characteristics of the load sequence,the SVMD(successive variational mode decomposition)is adopted and the improved sparrow search algo-rithm(ISSA)is invoked to optimize the parameters of the model,so as to avoid the parameter adjusting from falling into local optimum.The prediction effects and economic benefits of the different models are contrasted.The results show that the SVMD-ISSA-CNN-TGLSTM model has the best economic benefit,and its evaluation indexes RMSE(root mean square error),MSE(mean square error)and MAE(mean absolute error)are reduced by 35.7%,59.0%and 32.7%,respectively,in comparison with those of the ISSA-CNN-TGLSTM model,and all of the results are better than the other models,so its prediction effect is the best.关键词
供热负荷预测/逐次变分模态分解/改进的麻雀搜索算法/卷积神经网络/转换门控长短期记忆神经网络/空间提取能力Key words
heat load prediction/SVMD/ISSA/CNN/TGLSTM neural network/spatial exiraction capability分类
信息技术与安全科学引用本文复制引用
薛贵军,牛盼,谢文举,李水清..基于SVMD-ISSA-CNN-TGLSTM的供热负荷预测模型[J].现代电子技术,2024,47(11):131-139,9.基金项目
河北省自然科学基金项目(F2021209006) （F2021209006）
河北省高等学校科学技术研究项目 （）
