ABSTRACT
The operation of centrifugal compressors is limited by the surge instability at low mass flow rates. Being able to predict the surge limit during early compressor design stages would save manufacturers time and reduce the need for costly experimental testing. In this paper, we assess the mathematical model by Powers et al. [1] for its ability to predict the directional change in the surge limit when geometric dimensions are varied. The model is able to match the predicted directional change in surge limits from a data-driven model, and so shows good promise for being a usefultool.
