Chillers usually account for a major part of the overall energy consumption in commercial buildings. To improve the energy efficiency of chillers

we present an optimal chiller loading control strategy based on the best-performance line of the chiller. The best-performance line is created from historic operation data by using the moving window and data proliferation methods. The strategy is validated using an airport air-conditioning system in south China. The strategy is compared with sequencing control strategy. The results show that the proposed strategy shows a better performance with respect to energy saving. On typical days

the operating condition of the chillers can be optimized to save energy up to 8.2%.