A Model Predictive controller of a direct expansion air conditioning system for simultaneous temperature and humidity control

For direct expansion air conditioning systems, simultaneous control of indoor temperature and humidity has always been a tough challenge on account of the complexity of heat and mass transfer. To overcome this issue, in this paper, a controller based on model predictive control (MPC) method is developed for a direct expansion air conditioning system. First, the structure of direct expansion air conditioning systems is briefly introduced and the nonlinear state space model is established by mass and energy balances. By the Runge-Kutta method, the nonlinear model is directly used to build prediction model used in MPC controller. Then, the design of MPC algorithm is described in detail. The cost function is a positive definite quadratic function about indoor temperature and humidity. Finally, the simulation results demonstrate that the MPC controller can regulate indoor temperature and humidity to their respective desired values and enhance the dynamic quality, with shorter settling time, smaller the overshoot, compared to PID control strategy.