Comparative study of different control strategies for radiant ceiling panel system

Radiant ceiling panel (RCP) system has been increasingly adopted in recent years. However, its optimal control strategies have yet to be established. In previous studies, a grey-box model has been developed to predict the indoor thermal condition when using a RCP system. Based on this model, a control framework was implemented in MATLAB/Simulink, in which the inlet water temperature was regulated by a controller to maintain the indoor air temperature at a setpoint of 20 ℃. Different sampling period (1 min,10 min, 30 min and 1hour) were investigated with respect to system accuracy. Three control strategies- on off control, PID control, and model predictive control (MPC) were evaluated in terms of their performance on indoor temperature control and energy consumption reduction. For PID control, the gains were tuned by Ziegler–Nichols tuning method, and performance comparisons were made among P control, PI control and PID control. The results show that both PID control and MPC can effectively maintain the indoor temperature at the setpoint while consuming less energy consumption compared to on off control. In particular, MPC outperformed PID control by exhibiting fewer temperature fluctuations, owing to its ability to optimize control actions based on future state predictions.