Two loss decomposition methods for high-load turbine cascades

This study addresses the flow loss decomposition problem in high-load turbine cascades and proposes two loss decomposition methods based on different principles. The influence of cascade geometry parameters on the loss distribution is analyzed. Method I is based on flow structure characteristics and decomposes friction, shock wave, wake, and mainstream loss through flow field discretization and shock wave identification technology. Method II is based on flow partitioning, dividing the cascade passage into different loss-dominated regions and calculating the loss in each region through the mass-averaged parameters. An orthogonal experimental design is adopted, with the trailing edge radius, outlet Mach number, and trailing edge wedge angle as variables. The results of loss decomposition by the two methods are compared. The results show that the total loss trends calculated by the two methods are consistent, but the total loss calculated by Method II is generally lower than that of Method I. The main difference comes from the friction loss. The range analysis shows that the trailing edge radius has the greatest influence on the total loss, followed by the outlet Mach number, and the trailing edge wedge angle has the least influence. In addition, Method I is more suitable for refined analysis of shock wave position sensitivity, while Method II has higher calculation efficiency, but the accuracy of extracting boundary layer parameters needs to be optimized. This study provides theoretical support for the loss mechanism analysis and optimal design of high-load turbine cascades.