基于多元线性回归模型的建筑能耗预测与建筑节能分析  

Energy Consumption Forecasting and Energy Saving Analysis of Urban Buildings Based on Multiple Linear Regression Model

作　　者：樊丽军[1] 

机构地区：[1]山西大学工程管理系,山西太原030000

出　　处：《湘潭大学自然科学学报》2016年第1期123-126,共4页Natural Science Journal of Xiangtan University

基　　金：山西省科技厅项目(2015041011-3)

摘　　要：针对城市建筑能耗的节约与有效利用,提出一种基于多元线性回归模型(MLRP)的建筑能耗预测与建筑节能分析模型.以天然气和电力为能耗目标,将建筑类型、建筑年代、占地面积和居住人数等参数作为输入特征,利用多元线性回归模型分析出对能耗具有显著性影响的因素,并预测整个区域的能耗.另外,通过该预测模型,可以评估实施改善措施后建筑的节能潜力.实验给出了各种场景下的建筑节能潜力,分析结果表明,提出的预测模型能够精确预测区域能耗.For the issues that the saving and effective use of urban building energy consumption,a model of energy consumption forecast and energy saving analysis of urban buildings based on multiple linear regression model（MLRP）is proposed.This paper takes the natural gas and electric power as the energy consumption target,the building type,building age,floor area and number of residence as input characteristic parameters of multiple linear regression model,to analyze the factors which have a significant influence on energy consumption,so as to forecast the energy consumption of the whole region.In addition,it can evaluate the energy saving potential of the building after the implementation of the improvement measures by the prediction model.The experimental results show that the model can predict the regional energy consumption accurately,and give the building energy saving potential of various scenarios.

关 键 词：能耗预测 建筑节能分析 多元线性回归模型 显著性影响因素 

分 类 号：TU111.195[建筑科学—建筑理论]