|
| Shape Optimization Design of Intake Volute for an Axial Flow Compressor |
| DOI：10.13949/j.cnki.nrjgc.2025.03.010 |
| Key Words:intake volute secondary flow loss elliptic basis function（EBF） neural network multi-objective optimization deflector |
|
| Hits: 85 |
| Download times: 8 |
| Abstract:An aerodynamic optimization system for axial compressor intake volute， which included parametric modeling， mesh division， numerical calculation and computational fluid dynamics（CFD） post-processing， was built based on Isight optimization platform. After the intake volute was parameterized， the optimal Latin hypercube sampling was used to obtain uniformly distributed sample points in space. The total pressure loss coefficient and the velocity unevenness of outlet surface were taken as the objective functions. The objective function values of all sample points were obtained through numerical calculation， and the elliptic basis function （EBF） neural network proxy model was established. Then the non-dominated sorting genetic algorithm-Ⅱ（NSGA-Ⅱ） with elite strategy was used to seek the optimization of the objective function.Three optimization results M1， M2 and M3 were selected on the Pareto front and compared with the original model. The calculation results show that the total pressure loss of M1， M2 and M3 and the exit velocity inhomogeneity of M1， M2 and M3 are reduced， and the secondary flow of M1， M2 and M3 is contained. After optimization of the modeling coupling deflector， the total pressure loss of the volute is further reduced， the outlet flow field is more uniform， and the aerodynamic performance of the volute is further improved. |
| View Full Text View/Add Comment Download reader |