Empirical model for evaluating power consumption of centrifugal chillers

Abstract

A novel steady-state empirical power consumption model using water pressure difference, instead of water flow rate or cooling capacity, as one of the input variables is proposed for centrifugal chillers of a cooling plant with multiple parallel chillers. It is possible to evaluate the energy consumption and performance of a chiller without measuring the chilled water flow rate, which is difficult to measure directly and accurately in many on-field operating chiller plants. The on-field measurement and analysis indicates that the accuracy of the proposed model is slightly superior to that of the four empirical models quoted from literature—Yik model, Braun model, Comstock model and Chang model. When the internal data is considered, the coefficient of variation of root-mean-square error (CV) of the novel model for the chillers 1–3# are respectively 2.51%, 3.48% and 2.67% with the vast majority of relative errors being ±8%, ±10% and ±10%, respectively. When external data are adopted, the proposed model is obviously superior to the other four models with CV values for chillers 1–3# respectively being 4.94%, 7.79% and 5.98% and the vast majority of relative errors being ±8%, ±12% and ±10%, respectively. This result is acceptable for engineering applications, such as fault detection and diagnostics (FDD), building energy audit, energy conservation management and operation management of chiller plants.