JAUE2022-099: Study on Energy Supply Framework for Buildings Based on Multi-Energy Complementary Theory

Abstract

With the improvement of urban electrification level, the building power consumption continues to increase, leading to the further widening of the peak valley difference of the power system and the deepening of the contradiction between supply and demand. In view of this problem, based on the theory of multi energy complementarity, the study proposed a power supply framework for office buildings that integrated electric vehicles （EV）, photovoltaic power and municipal power. Based on the actual monitoring data, the performance of three machine learning algorithms in predicting building energy consumption was compared, and the appropriate machine learning algorithm was selected. The travel parameters of EVs are analyzed, and the charge and discharge model of EVs was established. Based on the radiation intensity, temperature and photovoltaic panel parameters, a photovoltaic power generation model was built. The office building of Beijing University of Technology was used as an application case. The results showed that the frame could fully meet the normal operation of office buildings. Under random conditions, EVs, as mobile load storage resources, had high schedulable potential. In addition, the random forest algorithm performed better in short-term building energy consumption prediction