Research Of Thermal Comfort Predict And Control Algorithms Of The VAV Air Conditioning

The air conditioning system is a tool for indoor environment adjustment.In the 21 st century,people’s requirements for indoor environment are getting higher and higher,However,the existing air-conditioning system cannot meet people’s requirements for indoor comfort and energy saving,this paper proposes a thermal comfort model air-conditioning control system,which makes the air-conditioning system more humanized and solves the problems of discomfort and energy saving.This paper mainly studies the prediction of thermal comfort model and the control algorithm of VAV air conditioning system.Through the analysis and research of four kinds of indicators for evaluating thermal comfort,it is concluded that PMV is selected as the evaluation system of thermal comfort,Analyze the main factors that affect the output of PMV indicators,are Air temperature,Air velocity,Air humidity,Average radiant temperature,Human metabolism,Clothing thermal resistance,Among them,air temperature and air flow rate are the two factors that have the greatest impact on thermal comfort,and are also two factors that are easy to control.The PMV index is a mathematical model composed of multiple environmental factors and individual factors,resulting in the complexity and nonlinearity of the PMV index model.It is necessary to iterate continuously when calculating the PMV index,and the current PMV index cannot be measured,It will cause errors in the application of air-conditioning real-time control system.This paper uses the RBF neural network to predict the PMV index model,but the RBF network has problems such as slow convergence and low prediction accuracy,which is not conducive to future real-time control;Faced with the predicted problems,Use PSO algorithm to improve neural network parameters and establish PSO-RBF method prediction,In order to improve the accuracy of the model again,Aiming at the premature convergence of the standard PSO algorithm and the weak local optimization ability,The method of improving PSO algorithm inertia weight,acceleration factor and other parameters is used to better optimize the RBF network,and establish a PSO-RBF network prediction model.This article addresses the time-varying,nonlinear,and hysteresis problems of air conditioning systems,proposed control algorithm BP-PID control as the control of thermal comfort,And designed the BP-PID controller,Because the BP neural network has problems such as slow convergence speed and local extreme values,this paper chooses the improved PSO algorithm to correct the weights and thresholds of the neural network parameters,and establishes the PSO-BP-PID control,The simulation of the three controllers shows that the improved particle swarm optimization neural network PID control effect is more stable.At the same time,the thermal comfort terminal control model of the air conditioner was established,and the simulation showed that the thermal comfort control effect met the comfort requirements.