煤气与热力2025，Vol.45Issue(2)：78-84,7.
基于ISSA-BPNN模型的城镇燃气日负荷预测
Urban Gas Daily Load Prediction Based on ISSA-BPNN Model
肖荣鸽 1夏海平 1李雨泽1
作者信息
- 1. 西安石油大学石油工程学院,陕西西安 710065
- 折叠
摘要
Abstract
An improved sparrow search algorithm to op-timize BP neural network(ISSA-BPNN)model is pro-posed for gas daily load prediction to address the is-sues of low accuracy and susceptibility to local optima in the prediction process of a single neural network model.The prediction accuracy of the model combin-ing BP neural network and optimization algorithm is higher than that of a single BP neural network.Opera-tions such as selection,crossover and mutation can im-prove population diversity and avoid premature conver-gence of algorithms.The average absolute percentage error of ISSA-BPNN model is 0.039 3,which has the highest prediction accuracy and longest time consump-tion in the combined model.关键词
燃气日负荷/BP神经网络/改进的麻雀搜索算法/负荷预测Key words
gas daily load/BP neural network/im-proved sparrow search algorithm/load prediction分类
建筑与水利引用本文复制引用
肖荣鸽,夏海平,李雨泽..基于ISSA-BPNN模型的城镇燃气日负荷预测[J].煤气与热力,2025,45(2):78-84,7.
