Analysis of Energy Consumption and Energy-saving of Residential Buildings Based on LM-BP Algorithm

In order to improve the accuracy of statistics and analysis of building energy consumption and energy saving, the paper chose Hebei residential building as the object, analyzing their energy consumption and energy saving by means of LM-BP algorithm. The LM-BP algorithm could establish the response relationship between input and output, and the fitting effect is directly related to the precision of the model. So the practical physical phenomena must be quantified and normalized by the impact factors, which combines local convergence of Gauss-Newton Method with global characteristics of Gradient Descent Method to achieve the analysis of buildings energy consumption data. Through simulation, the algorithm can achieve the effect of heating, ventilation, orientation and building shape on residential building energy consumption, so as to promote the energy saving analysis.