基于多神经网络的综合能源系统用户侧冷热负荷预测模型  

Prediction Model of User Side Cooling and Heating Loadof Integrated Energy System Based on Multi Neural Network

作　　者：张文栋 刘子琨 ZHANG Wendong;LIU Zikun(Wuling Power Corporation Ltd.,Changsha 410000,China)

机构地区：[1]五凌电力有限公司,湖南长沙410000

出　　处：《微型电脑应用》2023年第9期210-214,共5页Microcomputer Applications

摘　　要：为了降低用户侧冷热负荷预测结果与实测值之间的相对误差,提出基于多神经网络的综合能源系统用户侧冷热负荷预测模型。针对综合能源系统能量流动形式,分析用户侧冷热负荷影响因素,提取负荷计算主要参数;结合BP网络、RBF网络和小波神经网络,优化多神经网络预测方法,完成综合能源系统用户侧冷热负荷预测模型设计。算例分析结果表明:该模型应用下,预测结果与实测值的平均逐时相对误差为16%,可实现综合能源系统用户侧冷热负荷的精准预测。In order to reduce the relative error between the user side cold and hot load prediction results and the measured value,a user-side cold and hot load prediction model based on multi-neural network is proposed.For the energy flow form of the integrated energy system,the influencing factors of user hot and cold load is analyzed,and the main parameters of load calculation are extracted.By combining BP network,RBF network and wavelet neural network,a multi-neural network prediction method is constructed,and the design of user hot and cold load prediction model of the integrated energy system is completed.The example results show that the average relative error between the prediction result and the measured value is 16%,which can realize the accurate prediction of the hot and cold load on the user side of the integrated energy system.

关 键 词：多神经网络 综合能源系统 用户侧 冷热负荷 预测模型 

分 类 号：TU831.2[建筑科学—供热、供燃气、通风及空调工程]