증기압축식 냉동시스템의 상태공간 모델링 및 강인제어

Abstract

Recently, compressor variable speed(CVS) system has been used for precise control of vapor compression refrigeration system. This system is Multi-Input Multi-Output(MIMO) system that control target temperature and superheat at the same time. And a merit of the system are to save the energy but a demerit is hard to install the system equipments. Various control methods are studied for design the MiMo control system such as PID control based on transfer function, fuzzy control, fuzzy PID control, optimal control based on the state space model and so on. Demerits in PID controller is that control at that time of target temperature and superheat is hard. Many researchers was trying to improve the interference effect between target temperature and superheat. 4 control variable at least should be considered for control of fan motor in condenser and evaporator. Therefore, Application of optimal controller based on the state space model is desirable for MIMO controller design. Dynamic modeling is needed for design the optimal controller. This paper deals with practical low order linear modeling and LQG controller design in state space for optimum control with robustness of vapor compression refrigeration system. At first, a low order linear model which is suitable for controller design was derived from partial differential equations of heat exchangers by considering dominant state variables for output variables and applying some assumptions. Next, LQG controller with state observer using Kalman filter is designed to obtain the system robustness against input noises. Computer simulations are conducted to prove the validity of the proposed controller with simplified practical model. Simulation results show that the proposed controller can achieve fairly good robustness against input noises.