Study On Methodology For Forecasting Residential Energy Consumption In Chongqing

People's demands for their life quality and living environment are rising with the developing of economic and science technique. The energy structure is closely related to the improvement of environment, and the energy consumption for living has greatly improved the people's living quality. Energy consumption is affected by factors from China and abroad and is a complex nonlinear system. The traditional linear predict method has advantages such as simple and easy explaining, but it can't deal with the prediction of complex nonlinear dynamic system. Neural network is a nonlinear dynamic system, which has strong nonlinear-mapping ability and strong robust and forgiveness, it has special advantages for nonlinear system problems, and it's very fit to deal with the prediction and decision of nonlinear economic system.Then the paper analyzes the characteristics of Chongqing residential buildings in geography environment, national policy and self-conditions, generally analysis the present situation of the residential buiding energy consumption of Chongqing, and put the indicators which affecting the energy consumption in Chongqing residential buildings into 14 factors by analyzing the characteristics of Chongqing residential buildings, and establishing the index system of BP neural network prediction model.This research brings the wildly used GM (1, 1) prediction model and neural network theory into the modeling of the prediction of the energy consumption of Chongqing residential buildings, and get the prediction model by using the original gray theory GM (1, 1) model based on the DPS data system, then make the multi-factor BP neural network prediction model of Chongqing residential building energy consumption by using the Cshap language based on the SQL server 2005 platform. The factors of BP neural network prediction model of Chongqing residential building energy consumption includes relative moisture, population, incomes, total social retai consumption , the total fixed capital invested , urbanization rate, GDP, actual residential area, gas using rate, urban employer number, the architechture GDP, the accumulate money, which show the actual economic significance. The results of the prediction model have good precision and fitting results, and fit the developing trend of the actual data of energy consumption of Chongqing residential buildings. The results can help making the methods and standards of energy efficiency of Chongqing residential buildings, guide the industry of Chongqing residential buildings, optimize the structure of Chongqing residential buildings, and give suggestions to the methods of energy efficiency of Chongqing residential buildings from the prediction of the energy consumption of Chongqing residential buildings.Finally this paper analyzes the present situation of Chongqing's residential buildings, point out the problems of the energy consumption of residential buildings, and get the energy efficiency rates of different envelop conditions by using energy efficiency simulation software, to give a reference data for the energy efficiency methods.