Research On Data-Driven Modeling And Optimization Control Of Air Conditioning Chilled Water System

The central air conditioning chilled water system is the main component of the central air conditioning system,which realizes intelligent and low energy consumption operation and is one of the main bottlenecks in the development of intelligent building technology.The central air conditioning chilled water system often consumes a lot of building energy due to difficulties in modeling indoor cooling loads,mismatched temperature control with pedestrian flow,and other issues.Therefore,this article focuses on the construction of a compressed central air conditioning chilled water system experimental platform,data-based analysis and modeling of central air conditioning chilled water cooling capacity,and optimization control of the chilled water system.Intelligent optimization algorithms,PID control,and other technologies are utilized to achieve fuzzy modeling and optimization control of the central air conditioning chilled water system,thereby reducing energy consumption.The main research content of this article is as follows:(1)Based on the operating mechanism of evaporators,this paper establishes an Adaptive Network based Fuzzy Inference System(ANFIS)model for evaporators based on clustering algorithms.In this paper,the correlation coefficient is used to analyze the data variables that affect the heat transfer of the evaporator,and the Fuzzy clustering and subtractive clustering methods are used to cluster the data under each working condition of the evaporator model identification to reduce the complexity of the model.The mass flow of refrigerant,the saturated temperature of evaporator,the saturated pressure of refrigerant,the inlet temperature of evaporator and the inlet pressure of evaporator are established as inputs,An evaporator model that outputs heat exchange from the evaporator.Comparative experiments have shown that this method can effectively predict the heat transfer of evaporators with a small amount of data.(2)A modeling method for central air conditioning chilled water based on improved ANFIS algorithm.This paper uses particle swarm optimization algorithm and Differential evolution to optimize the antecedent and consequent parameters of the ANFIS model combined with Fuzzy clustering,and then takes the central air-conditioning chilled water system as the research object,determines the input and output variables of the central air-conditioning chilled water system model through correlation analysis,and uses Mean squared error(MSE)as the error evaluation index of intelligent algorithm optimization.The comparative experiments show that the ANFIS central air-conditioning chilled water system based on the combination of Differential evolution,particle swarm optimization algorithm and Fuzzy clustering has significantly improved in adaptive parameter adjustment,high-precision prediction and ease of application,and can significantly reduce the sample deviation value.(3)Control strategy based on social particle swarm optimization algorithm to optimize PID parameters.In response to the problem of low accuracy in temperature difference control of chilled water supply and return in central air conditioning,this paper uses social particle swarm optimization algorithm to optimize PID parameters to achieve temperature difference control of chilled water in central air conditioning.The least squares identification method is used to establish a transfer function model between chilled water pump frequency and supply and return water temperature difference.A matching chilled water system PID control is built in Simulnk Model for joint simulation,The simulation and experimental results show that the PID control parameters optimized based on social particle swarm optimization algorithm have good control effects in the central air conditioning chilled water system.