Performance optimization of a transcritical CO2 refrigeration system using a controlled ejector

An on-line optimal quasi cascade controller for an ejector with variable nozzle throat area is proposed to improve the operating performance of the transcritical CO2 ejector refrigeration system. The optimal gas cooler pressure is tracked in real time by the controller including a tracker and a predictor. Using the system dynamic model, the dynamic responses of the system performance and ejector efficiency under variable nozzle throat area are first analyzed. Then the parameters of the tracker and predictor are determined by simulation respectively which exhibits a good dynamic characteristic with an acceptable settling time. Besides, the controller presents a good robustness under variable compressor speeds and mass flow rate of cooling water. Furthermore, the system performance is actually increased to the maximal value by the controller even at the variable operating conditions. Finally, the optimal controller is verified by experiments to be an effective way to improve the system performance automatically.