The Optimal Pre-cooling Time And Shutdown Time Forecast For Air Conditioning System

To realize the grand goal of socialist ecological construction.The state has formulated goals to speed up ecological construction,solve energy consumption problems,vigorously safeguard the green environment and promote energy conservation and emission reduction in the 19 th National Congress of the Communist Party of China.As a major energy consumer in the society,construction energy use accounts for about 60% of the total,and more than half of that is spent on air conditioning.It can be seen that the energy-saving potential of air conditioning is huge.First of all,we found that the human comfort was not guaranteed when the staff entered the room on the research of the air conditioning system under the traditional control system.Then we found that the ordinary pre-cooling air conditioning system guarantees the comfort of people to a certain extent,but pre-cooling room temperature affected by indoor structure as well as outdoor environment is great,even waste a lot of energy.Considering the disadvantages of the above two kinds of control mode,we propose an optimal control method for turning on and off the air conditioning to solve the indoor comfort and make full use of the thermal inertia of the building and air conditioning system.The outdoor temperature is much lower than the indoor temperature before turning on the air conditioning system during the transition season,and the indoor temperature can be maintained for some time after the system is shut down.Therefore,we put forward the improved all-new air pre-cooling air conditioning power on control method and the shutdown control method to ensure indoor comfort.And we analyzed the mathematical model of the system components,established the TRNSYS simulation model and verified the reliability of the system.Based on the model,the best pre-cooling and shutdown time was obtained for 25 pre-cooling days in xi’an during the summer period,and the factors affecting the optimal pre-cooling and shutdown time were determined.After comparison,it was found that the Elman neural network was most suitable for the prediction of the optimal pre-cooling and shutdown time of the air conditioning system.Take TRNSYS simulation data as input parameters,empirical formula is used to determine the network hidden layer neurons number range,on the basis of using compare different network error and convergence speed,further determine the precise number of neural network hidden layer neurons of air conditioning pre-cooling and shutdown time.Use the network and simulated data to predict the optimal pre-cooling and shutdown time.The correlation coefficient method and root mean square error method were used to evaluate the quality of the network.Analyze the energy consumption of air conditioning pre-cooling and shutdown control strategy,and compare energy consumption with the traditional control method.Results show that,the correlation coefficient of the established Elman neural network for pre-cooling and shutdown prediction of air conditioning was 0.69 and 0.66 respectively,and the correlation was strong.The mean square error is 0.12 and 0.012 respectively,and the prediction deviation is small.It is concluded that using this optimization control method can save energy consumption by about 30% compared with traditional control.