Feedforward Variable Temperature Difference Control Of Central Air Conditioning Refrigerated Water System Based On Load Forecasting

With the rapid development of modern economy,serious environmental pollution has also been brought.The main reason for environmental pollution is that the total energy consumption of society is too high.As one of the most commonly used equipment,central air conditioning accounts for more than 40% of building energy consumption.Therefore,in order to respond to the national policy of energy saving and emission reduction,under the premise of satisfying comfort,energy saving optimization of central air conditioning must be carried out.At present,most central air-conditioners use feedback system,but because of its large inertia and lag characteristics,the cooling capacity of central air-conditioners is often excessive,resulting in waste of energy consumption.Therefore,in order to solve this problem,a feed-forward variable temperature control method based on load forecasting for central air conditioning chilled water system is proposed.Through the analysis of the characteristics of variable temperature difference control,the temperature difference between supply and return water at the next time is deduced by predicting the cooling load at the next time,and the temperature difference control of the chilled water system is adjusted by using the optimized PID controller.The experimental simulation is carried out with the help of MATLAB software.Experiments show that this method avoids the problem of too much cooling caused by the large inertia of the central air conditioning system,makes the system achieve the purpose of providing cooling capacity as needed,reduces energy consumption,and improves the quality of the central air conditioning system,which has a certain practical value.Specific research contents are as follows:(1)The influencing factors of central air conditioning load are analyzed,and the BP neural network load forecasting model is established.The BP neural network is improved by Bayesian regularization algorithm and genetic algorithm,and the simulation experiment is carried out in MATLAB.Experiments show that the improved BP neural network has the advantages of strong generalization ability and high prediction accuracy for the cooling load prediction of central air conditioning.(2)According to the process characteristics of chilled water system of central air conditioning,the transfer functions of fan coil unit,cooling room,temperature controller and load influencing factors are determined.According to the working principle of chilled water system,the mathematical model of chilled water system is analyzed and established.(3)In view of the shortcomings of traditional PID control,adaptive genetic algorithm is used to tune the parameters of PID.Through experiments,the control effects of traditional PID control,genetic algorithm optimization of PID control and adaptive genetic algorithm optimization of PID control on chilled water system are compared.Experiments show that the adaptive genetic algorithm optimizes the PID control to improve the comprehensive performance of the chilled water system and enhance the robustness of the system.(4)Through the analysis of variable temperature control method,the relationship between cooling load and temperature difference between supply and return water is obtained.The feed-forward variable temperature control of central air-conditioning chilled water system is carried out by predicting the cooling load at the next moment,and three optimized PID control methods are used to carry out simulation experiments.Experiments show that the three methods have better strain ability when the temperature difference between supply and return water in chilled water system changes,and the adaptive genetic algorithm is the best one.