Influencing factors regression analysis of heating energy consumption of rural buildings in China

The heating energy consumption in rural residential buildings is increasing in recent years. The 181 influencing factors which influence heating energy consumption mainly includes five parts: family basic information, rural residential building features, building envelope information, indoor air quality in winter and building heating energy consumption. Multiple linear regression analysis and logistic regression analysis were used to analyze the significant factors which affect rural building heating energy. Suitable and validated multiple regression model can use less variables to describe, explain and predict the heat energy consumption of rural residential buildings. Comparing different multiple linear regression models, one interactive exponential model which has goodness of fit, less predictive relative error and less influencing factors is optimal. This exponential model can be applied to predict heating energy consumption and annual heating energy consumption of per degree-days heating area of rural buildings. Logistic regression analysis can predict heating energy consumption from high, medium or low probability prediction classification and can evaluate the heating energy consumption level. Two regression analysis methods present a reliable, valid, and economical instrument for in-depth rural building energy saving research.