Research On Thermal Comfort Control Method Of VAV Air Conditioning Based On Bird Swarm Algorithm

Air conditioning system is an important tool to regulate the indoor environment.With the improvement of indoor comfort requirements,aiming at the characteristics of the existing air conditioning systems,which are not particularly satisfactory for the comfort and energy saving,a method of referring the thermal comfort indexes to the air conditioning system is proposed,which realizes the transfer of air-conditioning action point.Air-conditioning action point is transferred from the air to human body,which fully reflects the "people-oriented" concept,making the air-conditioning system more humane.Based on the review of thermal comfort theory,the effect of six influencing variables on PMV was analyzed by MATLAB,and effect of air temperature and air velocity was greater,which provided the basis for determining the control variables of the system.In view of the complicated calculation and heavy workload of PMV,the paper adopts the method of establishing PMV prediction model,and realizes the PMV control can be applied to real-time control system.Based on the analysis of the advantages and disadvantages of neural networks,the BP neural network prediction model of PMV is established.In order to solve the problem of slow convergence speed and low prediction accuracy of BP neural network,this paper proposes a method of Bird Swarm Algorithm(BSA)to optimize initial weights and thresholds of BP neural network.Then,the BSA algorithm is improved by adopting the Lévy flight method,which improves the shortcoming of slow convergence and weak local optimization ability of the BSA algorithm.Finally,the model of predicting PMV by using BSA-BP algorithm and LévyBSA-BP algorithm is established by using MATLAB.Simulation results show that PMV prediction model based on LévyBSA-BP algorithm has higher convergence speed and prediction accuracy than BSA-BP algorithm and BP neural network of prediction model.In this paper,aiming at the non-linear,time-varying and hysteresis characteristics of the actual air-conditioning system,the fuzzy PID controller is proposed and designed.Then,in view of the problem that the fuzzy PID parameters depend on the experience and a large number of test,the LévyBSA algorithm is used to optimize the fuzzy PID quantization factors and scaling factors to realize the online self-tuning of the PID parameters.Finally,MATLAB is used to simulate the basic PID,fuzzy PID and improved fuzzy PID.The simulation results show that the improved fuzzy PID control has less overshoot and stronger stability than the other two controllers,and the control effect is better.In this paper,the model of thermal comfort control of VAV air conditioner is established by simulink,and the simulation results are compared with the traditional temperature control.The results show that the thermal comfort control has better comfort and energy saving.