A multi-layer optimal chiller operation management framework

This work targets on the chiller plant operation management for a large campus building cooling load. In this paper a multi-layer optimal chiller operation management framework is proposed to operate various chiller units to meet the system cooling load while reduce the system operation cost. The first layer is the day-ahead 24-hour chiller operation planning layer which optimizes multiple chiller sequencing and chillers loading. The second real-time dispatching layer effectively addresses load forecasting uncertainties in real-time. The load uncertainty is solved hierarchically through two steps of approaches. An efficient and accurate system modeling framework for optimization is formulated with mixed-integer linear expressions to reduce the computational complexity. The effectiveness of this proposed management framework is demonstrated for an actual chiller plant operation in university campus. Simulation analyses validate that the proposed method can effectively address various levels of forecast uncertainty and reduce the operation cost.