A computational methodology for ejector design and performance maximisation

Abstract

Despite interesting potential for heat recovery and efficiency improvement that supersonic ejectors can bring to energy consuming systems, their successful use in industrial applications strongly depends on high performing designs. In order to reach this objective, an attempt is made to bridge the gap between available performance enhancement techniques and the complex physics of supersonic ejector internal flow structure. A new methodology, based on CFD simulations is thereby developed. It minimises the irreversible energy losses resulting from the formation of strong shock waves, the ultimate aim being to optimize ejector operation with a minimum number of simulation runs. The usefulness of this methodology is demonstrated on two different test cases showing under the same operating conditions, the performance of a base case ejector tested in the laboratory and an alternative design based on the proposed approach. The CFD model used for this purpose was previously developed and experimentally validated over a wide range of conditions. The results obtained in terms of entrainment and compression ratios show improvements of up to 29% on ejector performance.