Development of an Automatic Optimization code for Axial Compressor Blades using Genetic Algorithm

Abstract

Modern industrial axial compressor requires high performance for cost saving. The internal flow characteristic in axial compressor is very complex because of boundary layer separation and shock wave due to the adverse pressure gradient and high rotating speed. Because trial-and-error method is inefficient to improve the performance of the compressor, many researchers have developed the optimization method. In order to control the complex flow inside the rotating blade passage, the blade shape which directly effects the internal flow is the most important parameter in compressor design. In this study a two-dimensional automatic blade optimization process is established for the axial compressor cascade, NASA rotor 67. Genetic algorithm (GA) coupled with computational fluid dynamics (CFD) is applied as the optimization algorithm. Two cases of optimization are performed at two flow conditions; transonic and supersonic conditions. First case is the blade optimization for single objective: design point efficiency. Second is the multi-objective optimization of axial compressor cascade. The optimized blade shapes show increased efficiency and higher pressure ratio for each optimization process. All of the optimized results are compared with previous things through analysing blade profiles and results of CFD analysis such as pressure coefficient distribution, flow field and pressure contour.