Study On Energy Saving Of Air Conditioning Water System In Subway Station Based On Trim And Respond Control

In modern cities,traffic congestion,urban pollution,and people's difficulty in riding a car are becoming more and more difficult.The emergence of the subway has effectively improved these problems,and will bring about social and economic benefits.However,with the increasing cost of subway operation,the air conditioning system of the subway station is the big energy consumption of the whole subway station system,accounting for 30%-50% of the total electricity consumption,so the energy saving of the air conditioning system of the subway station is of great significance.Because of the large equipment selection,single control strategy,and dynamic change of station load in the ventilation and air conditioning system of the subway station,energy-saving measures are not in place,resulting in a great energy waste.It is necessary to analyze the load factors of the air-conditioning system and formulate a reasonable and feasible control strategy.In summary,we have studied from the following aspects:(1)Analyze of Factors Influencing Load of Air Conditioning System in Subway StationsThe construction of subway stations is different from that of above-ground buildings.The factors influencing the operating load of air-conditioning systems are complex,such as train piston wind,passenger flow,etc.These factors that affect the air-conditioning load in subway stations cannot be directly measured.The usual calculation methods are rough and with deviations.This paper proposed an area-based grey relational analysis algorithm to analyze the factors affecting the air conditioning load and built a TRNSYS simulation platform based on the ventilation and air conditioning training platform of a subway station in a certain university in Beijing,Based on the data generated by the simulation system,a grey relational model was established and its correlation was analyzed.Compared with other analysis variables,the experimental results show that the algorithm can more accurately analyze the correlation between the influencing factors and load.And the prediction model with correlation analysis is more accurate than the prediction model without correlation analysis.(2)Research on Air Conditioning Load Forecasting Model of Subway Station Based on GAW-LS_SVMThe main factor influencing the load on the air-conditioning system of a subway station is the human body heat load.Due to the instability of the passenger flow,the data samples used for modeling inevitably contain abnormal values,which greatly affect the quality of modeling.This paper proposed a weighted least squares support vector machine modeling method based on sample error.According to the characteristics of sample fitting error distribution,an improved weight rule is proposed to reduce the influence of sample error on the accuracy of modeling.This paper used a Seeker Optimization algorithm to optimize the parameters of LS_SVM.Compared with other weighting rules and methods without sample weighting,the results show that the proposed algorithm can better predict the air conditioning system load and the prediction accuracy is higher.(3)Energy-saving research based on Trim and Respond control strategyThe load of the air-conditioning water system in the subway station is designed according to the maximum capacity,but most of the air-conditioning system operates under partial load,resulting in a great energy waste.Based on the predicted load of the subway station,this paper studies a control strategy based on Trim and Respond control strategy,in which we take the pressure at the outlet pipe of the pump as the control variable and use the Trim and Respond rules to control the pressure value.Compared with the variable temperature control strategy,the experimental results show that the Trim and Respond control strategies save energy by 12% compared to the variable temperature control strategy.(4)Verification of Trim and Respond control strategy and energy saving effectiveness in subway training platformsTrim and Respond control strategy was applied to achieve optimal control of air-conditioning systems on training platforms.Based on the training platform operational data,a load forecasting model based on GAW-LS_SVM was established.In order not to verify the feasibility and energy-saving effect of the control strategy in this paper,fixed-frequency control and temperature-based control based strategies are applied to the training platform.Experiments show that the energy saving rate of pruning and response control strategy is 20.43%.