To design an aerodynamically efficient blade the distribution of entropy generation on the blade surface should be known. Having only knowledge of the integrated loss, makes the task of improving the efficiency of a blade extremely difficult. A method to predict the entropy generation rate in steady, two-dimensional, incompressible, adiabatic boundary layer flows is presented, which gives both the distribution and magnitude of the entropy generation rate. This prediction method is based upon five correlations which are used to determine the: 1. entropy generated in the laminar region; 2. entropy generated in the turbulent region; 3. location of transition; 4. length of transition; 5. entropy generated in the transition region. These are then used to predict the entropy generation rate on the suction surface of a turbine rotor blade at a moderate Reynolds number; comparisons are then drawn with past measurements. The aim is to develop a quick, simple and relatively accurate method for the prediction of entropy in the boundary layers of turbomachines, although the method is not confined to this application. The only information required to implement this prediction method is the boundary layer edge velocity distribution and the turbulence intensity. A benefit of this method is that it does not rely upon dissipative CFD predictions, which are both slow to use in a design process and not yet sufficiently trustworthy. The dissipation coefficient and entropy generation rate predicted for this test case compare well to experimental measurements, with the percentage difference between the integrated entropy measured and predicted being approximately 13%. However, the difference in the turbulent region is found to be as high as 30%.

This content is only available via PDF.
You do not currently have access to this content.