Research On Energy Saving Optimization Of Air Conditioning Chilled Water System Based On Load Forecast

Central air-conditioning can improve the indoor office and living environment,but it also brings huge energy consumption.Relevant data show that the current energy consumption of central air-conditioning in China can account for more than 40% of the energy consumption of the building industry.In central air conditioning,the energy consumption of the chilled water system is very large.Reducing its energy consumption through energy-saving optimization control plays an important role in reducing the energy consumption of the entire system,and accurate load forecasting can provide effective energy-saving optimization for the chilled water system.Therefore,this article takes the chilled water system as the object,and based on the air conditioning load forecast,the chilled water system is energy-saving and control optimization,so as to realize the energy-saving of the air-conditioning system.First of all,mathematical analysis is carried out for the two methods of chilled water temperature difference control-constant temperature difference control and variable temperature difference control.By comparing the energy saving potential of the two control methods under the same load ratio,it is finally decided to adopt the variable temperature difference control method as the refrigeration method in this article.How to adjust the water system.Then,a virtual model of the experimental room was constructed in TRNSYS,and the cooling season load required for the experiment was obtained through simulation.The Least Square Support Vector Machine(Least Square SVM,LS-SVM)model is used to predict the air conditioning load,and the best input data set of the model is determined by comparing the prediction effects of the LS-SVM model under different sets of input data.Aiming at the problem that the kernel parameters and regularization parameters are difficult to accurately determine when the LS-SVM model is used,an improved genetic algorithm(GA)based on an improved genetic algorithm(GA)is proposed to improve the LS-SVM model.Using the improved LS-SVM model to predict the air-conditioning load on two different forecast days,it is found that the improved LS-SVM model has a significant improvement in the prediction accuracy compared with that before the improvement,and the average relative error of the prediction results is from 2.08.% And 1.39% decreased to 1.43% and 0.70%.Then,the energy consumption model of the related equipment of the chilled water system is established by means of mathematical modeling.At the same time,the control transfer function of the chilled water system is constructed based on the temperature difference between the supply and return water.After completing the related modeling work,optimize the chilled water system with the goal of minimizing the operating energy consumption,and use the improved genetic algorithm proposed in this paper to optimize the chilled water system energy consumption and operating parameters under the corresponding constraints.,So as to obtain the optimal energy consumption of the chilled water system and the temperature difference between the supply and return water under the optimal energy consumption.Through comparison,it is found that the energy consumption of the optimized chilled water system is significantly lower than that before the optimization,and the energy saving rate of the optimized system can reach up to 9.61%.Finally,in order to ensure that the system quickly and stably enters a new maximum state when the optimal temperature difference setting value changes,this paper proposes a Model Free Adaptive Control(MFAC)algorithm based on improved GA.By re-setting the optimal temperature difference for three consecutive time periods in the forecast day,the control effect is compared with the traditional PID and the improved MFAC.The results show that the improved MFAC has better robustness and shorter adjustment time when controlling the chilled water system.