Energy Consumption Simulation of the Air Conditioning System in Large-Scale Buildings

The working environment of air conditioning system in large-scale building is very complex, and there is no significant linear relationship between factors affecting energy consumption and energy demand of air conditioning system. This study adopts a nonlinear regression model: ANN (artificial neural network) model as energy model of air conditioning system. Take outdoor temperature, categorical day-of-week variable, equipment efficiency and terminal load as input, energy demand as output. Use energy consumption data in 2011 for network training, and energy consumption data in 2012 to verify the reliability of model. Based on energy analysis, the operation condition and the characteristics of energy consumption of air conditioning system for large-scale buildings in Nanjing could be precisely represented.