The Research On Multivariable Compound Control Of Variable Air Volume Air Conditioning System

With the science and technology developing rapidly,intelligent building is applied in people’s work and life more and more widely.As an important part of intelligent building,central air conditioning becomes the research hotspots of many scholars and experts.Variable air volume air conditioning system becomes the important content in air conditioning field with its low energy consumption and flexible control.Therefore,it is of great theoretical significance and practical application value to study how to make the VAV system operate more efficient,energy-saving and comfortable.The actual air conditioning system as the background,this thesis completed the overall design of air conditioning system and the design of part of subsystems.With the purpose of saving energy and meet the requirements of indoor environment comfort,designed the temperature and humidity controller of VAV system,achieved decoupling control.By designing the temperature and humidity control system of air conditioning,as the air conditioning system is large-lagging,large-inertia,non-linear,it is difficult to achieve the desired control effect using conventional control methods.Therefore,neural network intelligent forecasting technology is adopted into the air conditioning control system.Predict the future output by analyzing the structural characteristics and identifying its mathematical model of air conditioning system.So that VAV air conditioning control system has better steady-state performance and dynamic performance,reduce energy consumption.With the improvement of people’s living standards,higher requirements of environmental comfort have been proposed.Not only appropriate temperature is required,humidity should also reach a certain target.Due to the coupling connection between temperature and humidity,they will affect each other in the adjustment process.Neural network decoupling controller is designed and achieved good results.The main work is as follows:(1)Based on engineering back ground,according to the control requirements of air conditioning system,the overall scheme of VAV central air conditioning system is designed,including process flow,duct layout and equipment selection,design of new fan control system and design of monitoring and management system.Pipe static pressure control scheme,air volume control scheme,air supply temperature control scheme,fresh air volume control scheme and terminal device control scheme are designed and applied in engineering successfully.(2)According to non-linear,large-lagging characteristics of the air conditioning system,neural network predictive controller model based on improved BP-PID is proposed,improved BP-PID control is combined with neural network predictive controller.The neural network predictor adjusts the control strategy on-line,optimizes control performance of the system.And the weights of neural network are optimized by genetic algorithm.The simulation results show that,the method can speed up the system response and improve the dynamic performance and steady state performance of the system.(3)In order to improve the comfort of the indoor air environment,using multivariable compound control of temperature and humidity.In order to solve the coupling problem between temperature and humidity,a decoupling control strategy based on neural network predictive control is proposed.This strategy combines predictive control with decoupling control to make full use of their respective characteristics.Not only to make the system have good control performance,but also to solve the coupling problem between temperature and humidity,so that air conditioning output accurate,non-coupling temperature and humidity values,to achieve better control effect.(4)Establish room mathematical model,use MATLAB software,obtain the simulation curve of RBF neural network predictive control based on improved BP-PID and the simulation curve of neural network decoupling control under different conditions.Through the comparison and analysis of simulation curve,it can be seen that the addition of the predictor makes the air conditioning system have better steady-state accuracy and dynamic performance.The addition of the decoupler solves the coupling problems of temperature and humidity,and meets the higher requirements of the air environment comfort.Has good robustness,and has wide application value.The strategy makes the system have good robustness and wide application value.