Study On Feedforward Decoupling Adaptive PID Control Policy For Variable Air Volume Air Conditioning System With Three-parameter Based On Modified Single Neuron PID Algorithm

The variable air volume air conditioning system（VAVACS）can meet the control requirements for indoor temperature by means of keeping the constant temperature of supply air and changing the flow rate of cooling/heating indoor supply air.Compared with the traditional constant air volume air conditioning system（CAVACS）,VAVACS is widely used in modern public buildings because of its advantages of reducing the energy consumption of fans and achieving temperature control in different areas.However,for the VAVACS with the process characteistics of multi-parameter,non-linearity and strong coupling effects among the main control loops,the problem of mutual interference between different control loops often occurs during the operation of VAVACS,which results in the unstable operation of the system and decreases the control quality for VAVACS.Thus,on the basis of analysizing three control loops with severely coupling characteristics for VAVACS and combining multi-variable decoupling control theory,fuzzy control algorithm and single neuron adaptive PID control technology,this paper propose a design scheme of feedforward decoupling adaptive PID control for VAVACS with three-parameter and a modified single neuron PID algorithm（MSNPIDA）to tune the parameters of adaptive PID controlers.By means of MATLAB tool,this MSNPIDA and the feedforward decoupling adaptive PID control systems for VAVACS with three-parameter are programmed and configured,respectively,and the simulations are carried out.The main research results in this paper are given as following.1.Based on the analysis of the three main control loops of VAVACS,the mathematical models of the flow rate of chilled/hot water(Q_CW/HW)and supply air temperature(T_SA),the speed of supply air fan（n₁）and the static pressure of supply air(P_SA),and the flow rate of indoor supply air(Q_SA)and indoor temperature（T_n）are set up,respectively,a matrix expression including the coupling transfer functions for these three control loops is constructed by the matrix analysis method.2.Based on the method of feedforward compensation,the transfer function matrix of decoupling compensator is constructed in order to eliminate the coupling effect among these three control loops.By the multivariable frequency design toolbox（MFDT）in MATLAB software,the simulations of the step responses of the feedforward decoupling control systems for VAVACS with three-parameter are performed.The results show that the feedforward decoupling compensator designed in this paper can eliminate the coupling effect among these three control loops.3.On the basis of the errors and the variant rates of supply air temperature(T_SA),the static pressure of supply air(P_SA)and indoor temperature（T_n）and the rules of fuzzy operation,this MSNPIDA can adjust the neural network weight valueω_i（i=1,2,3）and gain coefficientλof the single neuron,and the parameters K_P^*,K_I^*and K_D^*of three adaptive PID controllers are adaptively tuned,respectively,and obtain the corresponding optimal values.Meanwhile,with the help of MATLAB tool,this MSNPIDA is programmed and packged into the functional module.The related results from verification on classical functions and level control experiment of a water tank show that this MSNPIDA is reasonable and feasible.4.The feedforward decoupling adaptive PID control systems for VAVACS with three-parameter based on MSNPIDA are configured by MATLAB/Simulink tool and the numerical simulations are carried out under the operating conditions of air-conditioning in winter and summer.The results show that the corresponding control indexes of supply air temperature(T_SA),the static pressure of supply air(P_SA),and indoor temperature（T_n）can meet the relevant requirements of air conditioning technological process,and the corresponding tracking performance and anti-interference ability are better.5.Similarly,the coupling adaptive PID control systems for VAVACS with three-parameter based on MSNPIDA and the decoupling traditional PID control systems for VAVACS with three-parameter are configured,respectively,and the corresponding simulations under the operating conditioning of air-conditioning in summer are performed.By analysizing of the above results,it can be seen that the control performances of the proposed feedforward decoupling adaptive PID control method for VAVACS with three-parameter based on MSNPIDA are obviously prior to those of the other two control methods.