A semi-empirical modeling technique for predicting improved performance of water-cooled chillers used in building space cooling applications

This paper presents a semi-empirical modeling technique used to predict improved performance of water-cooled chillers used in HVAC applications. The model predicted the system's coefficient of performance (COP) and compressor total energy consumption as a function of condenser inlet-water temperature, evaporator outlet-water temperature, and evaporator load. The model was validated and demonstrated using experimental data from two case studies: a 675 kW (192-ton) chiller used at a high school in South Texas and a 10,550 kW (2,000-ton) chiller located at the physical plant of a university campus in Arkansas. The results showed that the average steady state measured and predicted COP as well as the average measured and predicted compressor energy consumption differed by no more than two percent. Simulations were run to identify improvements in chiller performance and opportunities to save energy.