Chilled water systems have large time delays and large inertia, and the traditional PID controller has a poor control effect. In this paper, an improved sparrow search algorithm is proposed to optimize the control of chilled water systems. Firstly, the random walk strategy was used to randomly perturb the sparrows to improve the searching ability of the sparrows. Then, a Gauss mutation was added in the iteration process of sparrows to enhance the local search ability. Finally, the values of the PID parameters as obtained by the above methods were substituted into the controller for simulation. The simulation results show that the method proposed in this paper improves the search accuracy of the sparrow search algorithm and effectively solves the problems of large time delays and large inertia in the chilled water system. The method in this paper took the least amount of time for the system to reach the steady state at only 12.75 s. The control effect of the proposed method was also better than that of the improved ant colony optimization algorithm. The rise time was 2.713 s, and the adjustment time was 4.95 s.

Background: Energy conservation has always been a major issue in our country, and the air conditioning energy consumption of buildings accounts for the majority of the energy consumption of buildings. If the building load can be predicted and the air conditioning equipment can respond in advance, it can not only save energy, but also extend the life of the equipment. Introduction: The Neural network proposed in this paper can deeply analyze the load characteristics through three gate structures, which is helpful to improve the prediction accuracy. Combined with grey relational degree method, the prediction speed can be accelerated. Method: This paper introduces a grey relational degree method to analyze the factors related to air conditioning load and selects the best ones. A Long Short Term Memory Neural Network (LSTMNN) prediction model was established. In this paper, grey relational analysis and LSTMNN are combined to predict the air conditioning load of an office building, and the predicted results are compared with the real values. Results: Compared with Back Propagation Neural Network (BPNN) prediction model and Support Vector Machine (SVM) prediction model, the simulation results show that this method has better effect on air conditioning load prediction. Conclusion: Grey relational degree analysis can extract the main factors from the numerous data, which is more convenient and quicker without repeated trial and error. LSTMNN prediction model not only considers the relation of air conditioning load on time series, but also considers the nonlinear relation between load and other factors. This model has higher prediction accuracy, shorter prediction time and great application potential.

Objective: In order to better understand the research results of AC load prediction and carry out new research, the Air Conditioning (AC) load forecasting method plays an important role in the energy consumption of AC. Method: This paper summarizes the methods of building AC load prediction, mainly from the impact factors of AC operating load and the methods of AC system operating load forecasting to introduce the current status of load prediction. This paper describes some studies on load influencing factors, compares the advantages and disadvantages of modeling methods for AC operation load prediction and points out the research direction of AC load forecasting. Results: The current research methods are summarized and analyzed. Traditional forecasting methods are no longer applicable to air conditioning systems. From the current research, combinatorial prediction has become a hot research object. This method combines two or more methods to reduce the prediction error and shorten the prediction time. Conclusion: This paper points out some shortcomings of the present research and the future research suggestions are given in the three aspects of sharing AC operation data, selecting the key factors of AC, and exploring the new methods.