Research On Energy Consumption Prediction And Quota Standard Of Urban Rail Transit Station

In recent years,urban rail transportation energy consumption problem is becoming a hot spot of attention.As an important part of urban rail transit energy consumption,energy consumption of urban rail transit station is an important part of energy conservation.The progress of The Times and the development of science and technology not only have higher requirements for the energy consumption of urban rail transit stations,but also provide more advanced technology.In order to save energy in urban rail transit stations,it is necessary to make clear the composition of energy consumption in existing stations,analyze the influencing factors of energy consumption,grasp the development trend of energy consumption,and evaluate the current state of energy consumption,so as to reduce energy consumption more reasonably and efficiently.This paper analyzes the main components of energy consumption in urban rail transit stations and the use of main energy consumption equipment.The influence factors of station energy consumption are discussed:first,the impact of single factor on station energy consumption is analyzed by combining qualitative and quantitative analysis.Then,based on the statistical processing of actual data,SPSS is used to conduct correlation analysis of the influencing factors to comprehensively consider the correlation degree between different influencing factors and the station energy consumption.In order to grasp the development trend of energy consumption,support vector machine is used to predict the energy consumption of station.Firstly,the outliers and normalization processing methods of station energy consumption data were proposed,and the principle of SVR and the solutions to existing problems in its application were summarized.The parameter optimization differences between CV-SVR and GA-SVR were compared with examples,and the results of the urban rail transit station energy consumption prediction model based on SVR were obtained.Compared with the prediction results of BP neural network,the prediction result R2 of GA-SVR is 0.89,which is obviously better than that of BP neural network,indicating that support vector machine has more advantages in short-term prediction of small samples.Because the urban rail transit station has the attribute of public building,this paper draws on the evaluation method of building energy consumption,and puts forward the quota standard of energy consumption of urban rail station based on energy consumption quota method.Based on the requirements of quota method,SPSS is used to test the distribution characteristics of energy consumption data of urban rail transit,and finally the energy consumption quota level table of urban rail transit is developed.According to the air conditioning season and the non-air conditioning season,three kinds of energy consumption quota levels,namely comfort energy consumption quota level,basic energy consumption quota level and luxury performance energy consumption quota level,are put forward respectively,so as to formulate the energy consumption evaluation standards for urban rail transit stations.Among them,the comfort energy consumption quota level is the energy saving target value of the recent station energy consumption,that is,the optimal value of energy consumption of urban rail transit station.Finally,based on the actual investigation and research results,this paper puts forward the key energy saving direction and measures of urban rail transit stations.