TY - JOUR
T1 - Investigation on aeroelastic characteristics of mistuned low-speed axial compressor rotor: Numerical methodology and optimization
AU - Fan, Chengwei
AU - Adjei, Richard Amankwa
AU - Ye, Zhouteng
AU - Cui, Jiahuan
PY - 2023
Y1 - 2023
N2 - In this paper, the aerodynamic and aeroelastic characteristics of a low-speed axial compressor were explored by using a multidisciplinary optimization approach. The numerical methodology utilized the parametric benefits of the free-form deformation and Gaussian process-based Kriging in predicting and optimizing the aerodynamic performance (total pressure ratio and isentropic efficiency) and statistical results of forced response (mean μ and standard deviation σ of amplitude magnification) with frequency mistuning. Based on this point, an optimization framework was proposed, coupled with in-house codes for calculating the aerodynamic damping associated with flutter and vibration amplitude of the forced response. Since only the rotor blade was optimized, the exit flow angle at the rotor outlet combined with blade equivalent von Mises stress and aerodynamic damping were chosen as constraints. The results indicated that the total pressure ratio and isentropic efficiency experienced increases by 0.4% and 1.7%, respectively at the point near peak efficiency, while simultaneously achieving a statistical reduction in amplitude magnification. Specifically, aerodynamic analysis revealed that blade twist, backward sweep and forward lean can contribute to the improvement of aerodynamic performance. However, it should be noted that the forward lean of the blade tip led to an increase of equivalent von Mises stress on the blade pressure surface. From aeroelastic analysis, data mining results implied a negative correlation between blade twist and aerodynamic damping, indicating that, an increase in the blade twist angle led to a reduction in aerodynamic damping. These findings highlighted one of the significant conflicts between aerodynamic and aeroelastic performance. Additionally, a strong linear correlation was observed between the mean and standard deviation of forced response, which indicated that a larger blade amplitude magnification increased its sensitivity to frequency mistuning.
AB - In this paper, the aerodynamic and aeroelastic characteristics of a low-speed axial compressor were explored by using a multidisciplinary optimization approach. The numerical methodology utilized the parametric benefits of the free-form deformation and Gaussian process-based Kriging in predicting and optimizing the aerodynamic performance (total pressure ratio and isentropic efficiency) and statistical results of forced response (mean μ and standard deviation σ of amplitude magnification) with frequency mistuning. Based on this point, an optimization framework was proposed, coupled with in-house codes for calculating the aerodynamic damping associated with flutter and vibration amplitude of the forced response. Since only the rotor blade was optimized, the exit flow angle at the rotor outlet combined with blade equivalent von Mises stress and aerodynamic damping were chosen as constraints. The results indicated that the total pressure ratio and isentropic efficiency experienced increases by 0.4% and 1.7%, respectively at the point near peak efficiency, while simultaneously achieving a statistical reduction in amplitude magnification. Specifically, aerodynamic analysis revealed that blade twist, backward sweep and forward lean can contribute to the improvement of aerodynamic performance. However, it should be noted that the forward lean of the blade tip led to an increase of equivalent von Mises stress on the blade pressure surface. From aeroelastic analysis, data mining results implied a negative correlation between blade twist and aerodynamic damping, indicating that, an increase in the blade twist angle led to a reduction in aerodynamic damping. These findings highlighted one of the significant conflicts between aerodynamic and aeroelastic performance. Additionally, a strong linear correlation was observed between the mean and standard deviation of forced response, which indicated that a larger blade amplitude magnification increased its sensitivity to frequency mistuning.
KW - Axial compressor
KW - Free-form deformation
KW - Multi-objective optimization
KW - Flutter
KW - Forced response
KW - Data mining
KW - Aeroelasticity
U2 - 10.1016/j.ast.2023.108735
DO - 10.1016/j.ast.2023.108735
M3 - Article
SN - 1270-9638
VL - 143
JO - Aerospace Science and Technology
JF - Aerospace Science and Technology
ER -