The optimization design of the actual fan/compressor rotors is a multidisciplinary problem. The aerodynamics and strength multidisciplinary optimization design of the fan/compressor rotors can improve the aerodynamic performance of rotors as much as possible under the condition of satisfying the structural strength requirements. However, the conventional multidisciplinary optimization of aerodynamics and structures will significantly increase the amount of computation cost and time period. Thus, it is difficult for conventional multidisciplinary optimization methods to be used in real-world engineering practices. In order to reduce the computational burden and time cost of the multidisciplinary optimization design, this paper firstly proposes a multidisciplinary optimization design method for rotors based on blade curvature constraints. In the optimization, the self-organizing map (SOM) method is applied to analyze and extract the constraint value of the blade curvature. And the blade curvature constraint value is used as a penalty function instead of the time-consuming high-fidelity FEM, which greatly reduces the computational cost and runtime of the multidisciplinary optimization. Besides, the Free-form deformation (FFD) method is adopted to realize the flexible deformation of the three-dimensional blade with the smallest number of design variables and polynomial chaos Kriging is used as a surrogate model for aerodynamic performance prediction in the optimization process. The method verification and mechanism analysis are carried out by studying a fan rotor. Data mining of FEM samples shows that, in the optimization design space, the maximum spanwise curvature of the blade is monotonically related to the maximum stress of the blade. And the maximum streamwise curvature, maximum spanwise slope, maximum streamwise slope, and maximum thickness of the blade have no related relationship with the maximum stress of the blade. Compared with the conventional multidisciplinary optimization design method based on the surrogate models of high-fidelity computational fluid dynamics (CFD) and finite element method (FEM), the time cost of the multidisciplinary optimization design method based on curvature constraints is reduced by 10 times, and the time cost of the multidisciplinary optimization is significantly reduced. The results show that, in terms of structural performance, the maximum stress of the aerodynamic optimization method without blade curvature constraints is 426 MPa, which exceeds the yield limit of aluminum alloy material (420 MPa), and the maximum stress of the multidisciplinary optimization design with blade curvature constraints proposed in this paper is 344 MPa. In terms of aerodynamic performance, the isentropic efficiency of the aerodynamic optimization method without blade curvature constraints and the method proposed in this paper are increased by 2.1% and 1.8%, respectively. Therefore, the proposed method reduces the maximum stress of the blade by 19.5% with little influence on the aerodynamic performance.
- Multidisciplinary optimization design
- Free-form deformation method
- Self-organizing map
- Curvature constraints