基于鲸鱼优化算法的高效制冷机房负荷预测研究  

Research on the load forecast of high-efficiency refrigeration room based on whale optimization algorithm

作　　者：潘亚男 崔红社[1] 刘高伟 PAN Yanan;CUI Hongshe;LIU Gaowei(School of Environmental and Municipal Engineering,Qingdao University of Technology,Qingdao 266525,China)

机构地区：[1]青岛理工大学环境与市政工程学院,青岛266525

出　　处：《青岛理工大学学报》2024年第6期118-125,共8页Journal of Qingdao University of Technology

基　　金：国家自然科学基金青年科学基金资助项目(52204039)。

摘　　要：为了准确预测建筑冷负荷,降低机房能耗,提出了基于鲸鱼优化算法(WOA)的BP神经网络预测模型。基于青岛办公建筑的实际历史运行数据,建立了BP神经网络、基于粒子群寻优的BP神经网络(PSO-BP)、基于遗传算法改进的BP神经网络(GA-BP)、基于鲸鱼优化算法的BP神经网络(WOA-BP)4种负荷预测模型,并比较了4种负荷预测模型的结果。研究表明,短期预测内,WOA-BP神经网络预测模型的最大百分误差为-15.76%,最小百分误差为-0.03%,平均绝对值百分误差为6.60%。与其他模型相比,WOA-BP模型具有更高的预测精度。In order to accurately forecast the cooling load of buildings and reduce the energy consumption of the refrigeration room,a BP(Back Propogation)neural network forecasting model based on the whale optimization algorithm(WOA)is proposed in this paper.Based on the actual historical operational data of office buildings in Qingdao,four load forecasting models were established,namely BP neural network,BP neural network based on particle swarm optimization(PSO-BP),BP neural network improved by genetic algorithm(GA-BP)and BP neural network based on whale optimization algorithm(WOA-BP).A comparative analysis of the results from these four load forecasting models was conducted.The research shows that in the short-term forecasting,the maximum percentage error of WOA-BP neural network forecasting model is-15.76%,the minimum percentage error is-0.03%,and the average absolute percentage error is 6.60%.Compared with the other models,WOA-BP model has higher forecasting accuracy.

关 键 词：建筑节能 鲸鱼优化算法 负荷预测 BP神经网络 

分 类 号：TK018[动力工程及工程热物理]