民居建筑室内空间热环境多点能耗精准预测  

Accurate Prediction of Multi-Point Energy Consumption in Thermal Environment of Indoor Space of Residential Buildings

作　　者：陈星星[1] 刘显成[1] CHEN Xing-xing;LIU Xian-cheng(School of Urban Construction,Yangtze University,Jingzhou Hubei 434000,China)

机构地区：[1]长江大学城市建设学院,湖北荆州434000

出　　处：《计算机仿真》2024年第2期51-55,共5页Computer Simulation

基　　金：2022年湖北省水利厅重点研发项目(HBSLK202209)。

摘　　要：在民居建筑室内热环境多点能耗预测时,若不能有效确定热环境下室内当前状态下舒适程度,会直接影响能耗预测精度。因此,提出考虑舒适度的民居建筑室内空间热环境多点能耗预测方法。根据室内状态建立民居建筑内热环境模型,采用主元分析法获取能耗评价指标,构建热环境下室内能耗指标评价体系,并利用指标建立评价模型,从而确定室内当前热环境能耗;将确定指标作为模型输入向量,利用RBF神经网络构建室内能耗多点预测模型;通过建立的模型实现对民居建筑室内空间多点能耗的精准预测。实验结果表明,所提方法开展室内能耗多点预测时,预测精度高、效果好。In the process of predicting the multi-point energy consumption of indoor thermal environments in residential buildings,if the comfort level of the current indoor thermal environment cannot be effectively determined,the prediction accuracy of energy consumption will be directly affected.Therefore,a method for predicting the multi-point energy consumption of the indoor thermal environment of residential buildings was proposed.Firstly,a thermal environment model was built according to the indoor state.Then,the principal component analysis was adopted to obtain the evaluation index of energy consumption.Secondly,an evaluation system of energy consumption in a thermal environment was constructed.Based on these indicators,an evaluation model was constructed to determine the current energy consumption.Thirdly,the determined indicators were used as input vectors of the model,and an RBF neural network was used to construct a multi-point prediction model of indoor energy consumption.Finally,the accurate prediction of multi-point energy consumption of indoor space in residential buildings was achieved through the model.Experimental results show that the proposed method has high prediction accuracy and good effect on multi-point prediction of indoor energy consumption.

关 键 词：民居建筑 室内空间 热环境 多点能耗预测 预测方法 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]