Study Of Energy-saving Operation Optimization For Chilled Water System Of Shopping Mall Based On Cooling Load Prediction

With the rapid development of large-scale integrated shopping malls, the problem of high erergy consumption in commercial buildings is getting outstanding. Because of long term part-load operation and lacking of effective control, the central air-conditioning system wastes lots of energy. The chilled water system is not only an important part of the central air conditioning system, but also has an important influence on the overall energy consumption of the system. It is of great significance to carry out the research of an optimal operation method based on online cooling load predicction for the chilled water system.In this paper, the chilled water system of an integrated shopping mall in Guangzhou is studied. According to the real-time operation datas, appropriate input parameters are selected to establish the air conditioning load predicction model through a SPSS correlation analysis. The air conditioning load predicction model is trained based on support vector regression, whose parameters are optimized by particle swarm optimization algorithm. The results shows that the prediction accuracy of SVR model based on PSO optimization algorithm is higher than that of BP neural network.The maximum relative error of the SVR model is within 10%, and the minimum relative error is within 1%.Energy consumption model of chillers and chilled water pumps are established by semi-empirical method, and an adaptive online identification method has been proposed to solve the problem of the energy consumption model changed with the time. The method is validated using real-time operation datas, and it shows higher precision and prediction precision compare to recursive least squares identification method and fixed forgetting factor recursive least squares identification method.The optimization objective function of the mall energy consumption of chilled water system is established, based on matlab platform, the chilled water system energy saving operation optimization parameters of the mall have been confirmed using genetic optimization algorithm, the result shows that the method is feasible, and the energy consumption reduces by 4.8%.The program of the online operation optimization of the chilled water system based on air conditioning load prediction in the shopping mall is proposed, and has been realized developing a program on the basis of hybrid programming using Matlab under C# programming environment. Experiments were taken to prove the energy saving effect of the optimal control. The results indicate that average energy consumption reduces by 4.4% using the optimal control method instead of the original control method.