Study On Two-degree-of-freedom Internal Model PID Control Policy Using Improved Multi-objective Particle Swarm Optimization Algorithm For Supply Air Temperature In CAVAHU

Constant air volume air conditioning system(CAVACS)and variable air volume air conditioning system(VAVACS)are two types of the main application of full-air conditioning methods in central air conditioning systems.In large public buildings,CAVACS is more widely used than VAVACS due to the advantages of relatively less initial investment of air-conditioning equipment,simple operation and management,easy maintenance,and stable air supply parameters,etc.In addition,with the rapid development of society and the increasing demand for indoor comfort degree(ICD),the stable supply air temperature for CAVACS is not only required to meet the ICD requirements,but also CAVACS operates in an energy-saving manner.However,traditional control strategies are selected as controlling supply air temperature for CAVACS,for example,a PID controller for supply air temperature(SAT-PIDC)is adopted to manipulate the flow rate of chilled and warm water flowing through constant air volume air handling unit(CAVAHU).The PID single-loop control method is simple and tuning parameters of SAT-PIDC is carried out off-line,which can not adapt to the characteristics of CAVAHU time delay,nonlinearity,and time-varying structural parameters for CAVAHU and usually leads to the problem of larger steady-state error,larger overshoot and longer regulating time,etc.Taking into account the above problems and the need to further improve the SAT control strategy for SAT,this paper proposes a two-degree-of-freedom internal model PID control system for supply air temperature(SAT-TDFIMPIDCS)by combining the air-conditioning process requirements for CAVAHU,the principles of internal model controller(IMC)and PID control technology.For tuning parameters of the two-degree-of-freedom internal model PID controller for supply air temperature(SAT-TDFIMPIDC),an improved multi-objective particle swarm optimization algorithm(IMOPSOA)is designed to obtain the optimal values of the corresponding parameter.With the help of MATLAB/Simulink software,this SATTDFIMPIDCS is configured and numerically simulated.The results verify the feasibility of the two-degree-of-freedom internal model PID control strategy for supply air temperature and IMOPSOA proposed in this paper.The main research tasks and contents of this paper are as follows:1.Firstly,based on the standard PSOA,and keeping other parameters constant,by exponentially decreasing the constriction factor,the structure model and calculation process of an improved single objective particle swarm optimization algorithm(ISOPSOA)are constructed by decreasing its constriction factor exponentially.Secondly,based on the typical multi-objective particle swarm optimization algorithm(MOPSOA),an improved multiobjective particle swarm optimization algorithm(IMOPSOA)is designed by adding the strategy of calculating the angle among the particles to select the global leader.ISOPSOA and IMOPSOA were applied to the classic functions of Rastrigin,Rosenbrock ZDT1 and ZDT2 to be verified,respectively.The results show that these two improved PSOAs are feasible.Compared with the standard PSOA and the typical MOPSOA,their convergence and diversity are greatly improved to some degree.2.Considering the air-conditioning process requirements for CAVAHU and the dynamic characteristics of the controller plant with supply air temperature,an SAT-TDFIMPIDCS is designed by integrating the principle of internal model controller,two-degree-of-freedom control method and PID control technology,and the models of cooling and heating coil in CAVAHU,measuring transmitter for supply air temperature,actuator to manipulate the flow rate of chilled and warm water and SAT-TDFIMPIDC are set up,respectively.In addition,according to the thermal adaptation model of the air-conditioned room,the setting value for supply air temperature(denoted as TSA,SET)is continuously output to this SAT-TDFIMPIDC to form a servo control for the supply air temperature.3.Considering the regulating time(denoted as ts),the absolute value of steady state error(denoted as ESS)and ITAE,min ITAE and min(ITAE,ts,ESS)are selected as a single objective fitness function for ISOPSOA and multi-objective fitness function for IMOPSOA,respectively.By running ISOPSOA and IMOPSOA,the optimal values of parameters of two-degree-offreedom internal model PID controller for the water tank level in the level control unit are obtained,respectively.The feasibility to tune parameters of two-degree-of-freedom internal model PID controller is verified through the numerical simulation and experiment of twodegree-of-freedom internal model PID control for the water tank level of for these two improved PSOA.4.With the help of MATLAB/Simulink tools,this SAT-TDFIMPIDCS is configured.Under the air condition conditions in winter and summer,the control effects of this SATTDFIMPIDCS and the Pareto optimal solutions of the corresponding SAT-TDFIMPIDC are numerically simulated and output,respectively.The results show that smaller overshoot,fast response,short adjustment time,steady-state error of zero and strong anti-jam are obtained for supply air temperature.5.For the same controller plant with supply air temperature in CAVAHU and the airconditioning process index,numerical simulations of single loop fractional PID and single loop PID for supply air temperature are carried out,respectively.Through the analysis of the results,it is seen that the two-degree-of-freedom internal model PID control method proposed in this paper is superior in the control indexes on supply air temperature.