Energy efficient operation based on root cause analysis for multiple-chiller plant

In large gas district cooling plant, chillers' performance bottleneck analysis is important not only for anomaly detection, but also for energy optimal operations. Currently many researches adopt linear or non-linear regression correlation analysis methods to detect chiller's bottleneck, however their works majorly focus on single chiller's anomaly factors analysis, which cannot be directly applied on operation optimization especially in large multiple-chiller plant, because single-chiller model cannot answer the questions of when and where the plant level's bottleneck happens. The present paper extends single-chiller anomaly analysis model RCA into a multiple-chiller oriented performance root cause analysis model RCA+, designs two energy optimal steam operation algorithms to meet cooling demand's rising and lowering, and verifies the algorithms' energy saving effectiveness by simulation. The simulation result shows that proposed optimal operation algorithms can implement about 11.6% higher steam efficiency than COP-based method for studied multiple-chiller plant.