The Optimization Of Thermal Comfort Prediction And Control Model For Indoor Thermal Environment

Nowadays, the indoor environment quality is demanded more and more strict. At the same time the people can most directly feel the indoor thermal environment. The primary means of adjusting the indoor thermal environment is the air-conditioning control system.The traditional air-conditioning control system controls the temperature of the indoor thermal environment by a single indicator, it is not really based on human thermal comfort.And the energy consumption of this control system is very high. So it is imminent to build up a comfort control system for indoor thermal environment. However, before constructing comfort control system, we must determine an evaluation standard of the indoor thermal environment. And then using this evaluation standard,make the indoor environment toward the state that people want to adjust and change. In summary, the results of this study are as follows:(1) This paper selects PMV model as evaluation standard of thermal comfort to evaluate the indoor thermal environment. Known by the PMV model, the calculation process of PMV value is very complex, and it has a significant time lag. To solve these problems, this paper use BP neural network to predict thermal comfort PMV and build PMV prediction model. Finally, through simulation experiments verify the feasibility of thermal comfort PMV prediction model for indoor thermal environment. And it can be used in real-time control of air-conditioning.(2) BP neural network algorithm has slow convergence rate and low prediction accuracy,and it easily gets into local optimal.So in this paper, PSO algorithm is adopted to optimize neural network algorithm. And improve PSO algorithm at speed renewal, inertiaweight and accelerating factor. Then an improved algorithm of PSO is proposed(IPSO).Using IPSO to optimize neural network algorithm,a real time PMV forecasting model of BP neural network with IPSO optimized is proposed. Through the experiment simulation,compared with the BP forecasting model and PSO-BP forecasting model, the forecasting model of this paper is faster and more accurate.(3) Since the traditional air conditioning control system is a complex system with nonlinear and large time lag. In this paper, the fuzzy control algorithm is used to establish the PMV thermal comfort fuzzy control model for the indoor thermal environment.Through the experiment simulation,it is concluded that the system is expected to achieve the desired effect.Compared with the traditional PID control system,the control effect of comfort control system used in this paper is better.