Research On Air Conditioning Load Forecasting In Subway Station Based On Practical Data

With the development of science and technology and the progress of society,China’s subway station construction has entered the era of rapid development.As a convenient and fast public transport mode,subway has become the main way for people to travel in the city.At present,the overall energy consumption of air-conditioning system in subway station is large and there is a phenomenon of energy waste.If the hourly air-conditioning load can be predicted according to the characteristics of subway station,the operation of air-conditioning system can be guided and optimized,and the energy waste caused by the mismatch of cooling supply and demand can be reduced,so as to achieve the purpose of energy-saving operation.In this thesis,the air conditioning load of a subway station in Beijing is studied.Firstly,the main influencing factors of subway air conditioning load are investigated and tested.In this thesis,the field test is carried out in the air conditioning season from June to August in 2019.The data of the thermal environment and air environment inside and outside the station are monitored and recorded,and the passenger flow data and the actual energy consumption data of the air conditioning system are obtained.The correlation between the parameters and the energy consumption of subway station is analyzed,and the main factors affecting the energy consumption of subway air conditioning are obtained.The analysis results show that the outdoor temperature,indoor CO₂concentration and passenger flow in the station are highly correlated with the energy consumption of subway air conditioning.Secondly,aiming at the problem that it is difficult to measure the natural infiltration wind in subway station at present,this thesis puts forward the method of measuring the infiltration air volume by using the artificial emission source CO₂tracer gas,and carries out the horizontal space and vertical space uniformity experiments on the distribution of CO₂air in the station to verify the feasibility of the method.The data are calculated by using the tracer gas steady-state difference method of two kinds of difference time,and the correlation analysis is carried out on the calculation results.The air volume calculation results show that the average calculated infiltration air volume of different difference time has little difference,and the infiltration air volume also has the characteristics of date type,and the infiltration air volume of different subway stations;the correlation analysis shows that the number of people in the station and the frequency of trains are different There are two factors that affect the infiltration air volume of subway station.Then,taking a day in August 2020 as an example,the actual test and monitoring data of outdoor temperature,passenger flow and natural ventilation infiltration air volume in the station are coupled,and the dynamic load of subway is calculated and analyzed according to the above test and calculation methods.The results show that the maximum dynamic load is less than the design load.According to the characteristics of subway air conditioning dynamic load,the control logic of subway air conditioning dynamic load is proposed based on the existing air conditioning system control mode.Finally,based on BP neural network,the dynamic load forecasting model of subway air conditioning is established,and the model is optimized by combining particle swarm optimization algorithm and genetic algorithm.Based on ga-pso-bp optimization forecasting model,the load forecasting in August 2020 is verified,which shows that the air conditioning forecasting model has faster forecasting speed and higher accuracy.