Study On Energy Consumption Prediction Model And Energy Saving Diagnosis Of Air Conditioning System In A Shopping Mall In Xi’an

With the rapid development of the city,shopping mall buildings also increase rapidly,and their energy consumption also increases rapidly.The average energy consumption per unit area of mall buildings is much higher than that of other public buildings due to the characteristics of heavy air-conditioning system load,high energy consumption for lighting,various types of energy-using equipment and dense personnel in the mall.The energy consumption in the air-conditioning system is about 5 times that of other types of public buildings.Studying the characteristics of energy consumption of air-conditioning in shopping mall buildings and finding reliable energy-saving measures are of great significance to energy-saving in shopping mall buildings.Therefore,the dissertation studies the air conditioning energy consumption prediction and energy saving diagnosis of a shopping mall building in Xi’an.The main contents are as follows:Firstly,the air conditioning system of a shopping mall in Xi’an was studied and the influencing factors related to its energy consumption were studied.The influencing factors of the energy consumption of the air conditioning system were divided into indoor influencing factors and outdoor meteorological factors.Indoor influencing factors include indoor passenger flow and historical data of air conditioning system;Outdoor meteorological factors include outdoor temperature,outdoor wind speed,outdoor humidity and outdoor solar radiation.These factors are input variables of energy consumption prediction model.Secondly,by analyzing the correlation degree between the input and output variables selected by the model,the PSO-BP neural network energy consumptionprediction model of air conditioning system in a shopping mall in Xi’an was established.Moreover,with the help of JMP software tool,pearson correlation significance was used to test,the influencing factors with low correlation degree were removed,and the influencing factors with high correlation degree were used as the input variables of the energy consumption prediction model.Markov chain is used for error correction to eliminate the process error caused by the PSO-BP neural network in the process of energy consumption prediction of air-conditioning system,so the established energy consumption model is optimized and more accurate prediction results are obtained.Simulation results show that the improved PSO-BP neural network energy consumption prediction model is compared with the improved PSO-BP neural network energy consumption prediction model.The maximum relative error of the improved PSO-BP neural network energy consumption prediction model is reduced from 33.799%to 10.967%,and the root-mean-square error is reduced from 16.14% to 3.06%,indicating that the improved PSO-BP neural network energy consumption prediction model is more accurate and more suitable for practical engineering applications.Finally,the energy saving diagnosis model of air conditioning system is established.The k-means clustering algorithm is used to select the data with good energy saving characteristics from the historical data as the sample data of the diagnosis model and establish the diagnosis model of energy saving.In this way,the abnormal energy consumption can be diagnosed,the data with poor energy saving characteristics can be analyzed abnormally,and the non energy saving time in the operation of the air conditioning system can be analyzed.Through this modeling method,we can find out the unreasonable situation in the operation of the air-conditioning system,and provide a reference for the management and operation of building energy conservation.The dissertation mainly studies the energy consumption prediction and energy-saving diagnosis of building air-conditioning systems in shopping malls,proposes an improved PSO-BP neural network prediction model and establishes an energy-saving diagnosis model,which can not only solve the energy consumption prediction and existing problems in energy-saving management,and have guidance and reference for energy consumption prediction and energy-saving management of air-conditioning systems of the same type of buildings in Xi’an.