Design of Cascade Model Using an Optimization Method

Abstract

If a linear turbine cascade experimental apparatus uses only few cascade model, it has several advantages in experiment, such as increasing the accuracy of experimental result by applying bigger turbine cascade model on the same apparatus without increasing compressor power, and easy access of optical devices within the cascade passage. However, it is difficult to make the same periodic flow conditions between blades as infinite cascade flow conditions. Generally, removing operating gas from walls of cascade experimental apparatus or adjusting tail boards located at downstream of cascade blade model are applied to make periodic flow conditions at cascade blades. In this study, sidewalls of turbine cascade experimental apparatus are designed to satisfy periodic flow conditions without any control of operating gas or tailboards. Two turbine blades are installed within a double-passage turbine cascade. Surface isentropic Mach number distribution on the blades is chosen as a responsible variable to match with the periodic flow condition on infinite cascade model. A gradient based optimization method is adopted for designing these walls. Computed result shows that it could obtain the same periodic flow conditions as infinite cascade model by modifying only sidewalls of double-passage turbine cascade experimental apparatus.