A LPV model-based chilled water temperature controller for HVAC systems

Abstract

In this paper, a linear parameter varying (LPV) model-based predictive control method is proposed to regulate the chilled water temperature of centralized heating, ventilating and air conditioning (HVAC) systems. First, a LPV model structure is proposed to identify the dynamics of the refrigerant loop under varying cooling load. By defining the scheduling variable to be the chilled water mass flow rate, the working conditions, under which the HVAC system is operating, can be quantified. Thereafter, a certain number of operating points across the whole operation range can be determined and the corresponding local linear models are built. Subsequently, a global LPV model is attained by interpolating all local models along with transition process data. With the aid of the model, a LPV-based model predictive control mechanism is implemented to control the compressor frequency in order to maintain the chilled water temperature in the presence of varying cooling load and external disturbance. Experimental tests on a pilot HVAC system demonstrate that the proposed control scheme can accurately capture the nonlinearity of the process and deliver satisfactory performance to maintain a desired chilled water temperature. Practical application: HVAC systems equipped with variable chilled water loop can save energy at part-load conditions. But modeling the dynamical characteristics of regulating chilled water temperature of such systems is challenging. This paper offers an LPV model methodology that can be implemented on a polite HVAC system to represent the dynamics of chilled water temperature. Based on the obtained LPV model, model predictive controller is subsequently designed to maintain the chilled water temperature. Specific procedures of LPV model identification and LPV-based predictive controller implementation have also been shown.