9 Aerodynamic Performance Analysis

An aerodynamic performance analysis is essential for nearly all aspects of axial-flow compressor aerodynamic design and application. There are basically two types of performance analysis techniques in common use: One-dimensional or mean-line methods analyze the performance along a mean stream surface. If applied to well-designed stages, the mean-line performance may be considered to be representative of the overall performance, at least near the compressor's design operating conditions. This approach is more questionable for off-design performance prediction. In those cases, blade incidence angle matching, blade loading levels, etc., may vary dramatically at different locations along the blade span, such that the mean-line performance is no longer representative of the overall performance. This is particularly true with respect to blade loading limits, blade stall and end-wall stall, which establish the compressor's surge limit. In off-design operation, the extremes in incidence angle and flow diffusion will almost always occur on the hub or the shroud contours. Some one-dimensional methods include approximate calculations at the hub-and-shroud contours to provide additional guidance to the user and to better evaluate tip clearance losses, shroud leakage effects and end-wall boundary layer blockage. The more general approach is to conduct the performance analysis for a series of stream surfaces from hub to shroud. These methods are referred to by various names, such as streamline methods, through-flow methods, streamline-curvature methods or three-dimensional methods. Hub-to-shroud performance analysis is a more accurate term, which is used in this book. It has been common practice to conduct hub-to-shroud performance analysis using the full normal equilibrium equation as described in Chapter 7. The longer computer times required and reduced reliability due to numerical instability for these techniques is probably the main reason that one-dimensional methods have continued to be used. But if the through-flow analysis offers the approximate normal equilibrium models described in section 7.6, the advantages of a hub-to-shroud performance analysis can be obtained with computation speed and reliability comparable to those of a mean-line method. Consequently, there is really no longer a need for one-dimensional performance analysis methods for axial-flow compressors.

The component parts of a hub-to-shroud performance analysis have been presented in Chapters 6 to 8. This chapter describes methods to integrate those component analyses into a hub-to-shroud performance analysis and suggests some useful features that can make it more effective. It also compares performance predictions from this performance analysis against experimental data for axial-flow compressors to demonstrate the merits of the procedures presented in Chapters 6 to 8.