Aeroelastic modeling and multi-objective optimization of a subsonic compressor rotor blade using a combination of modified NSGA-II, ANN, and TOPSIS

Abstract

A significant engineering challenge in the aerospace and energy sectors has been improving the aerodynamic and mechanical performance of subsonic axial compressor designs under high-loading conditions. A multi-objective optimization framework for a subsonic axial compressor is presented in this study through an integrated computational approach. Aeroelastic modeling is combined with advanced optimization techniques, including modified Non-dominated Sorting Genetic Algorithm II (NSGA-II), artificial neural networks (ANNs), and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). Complex fluid-structure interactions were evaluated using computational fluid dynamics simulations (CFD) and finite element analyses (FEA). Additionally, the Group Method of Data Handling (GMDH) has been used to predict performance across multiple design scenarios as a rapid surrogate model. During the optimization process, total pressure coefficients improved by 2.8 %, efficiency increased by 3.2 %, and maximum stress decreased by 8.9 %. By meticulously adjusting the stagger angle and tip clearance, these improvements resulted in reduced flow losses and a more uniform distribution of stress. As a result of this optimization, the stresses imposed on the blades at critical points are reduced, resulting in greater mechanical stability under diverse operating conditions. These improvements have also improved the compressor's aeroelastic performance. This reduces the noise levels during operation, making the compressors more efficient and environmentally friendly. Additionally, it also increases the lifetime of the compressor.