Multilevel Optimization of Rotor Bearing System With Dynamic Behavior Constraints

Abstract

The main purpose of this study is to investigate the multilevel optimization of rotor-bearing systems. The design variables include the shaft inner radius, the bearings stiffness, and the axial position of the bearings and disks. The design objectives are minimization of the shaft weight and transmitted force to the bearings. Constraints are placed on the critical speeds, the maximum shaft bending stress, and the maximum amplitude of the steady state response. In the multilevel optimization, three levels are considered and each level includes a single objective and/or multiobjective with various design variables and constraints. For each level, the method of feasible direction (MFD) is used. In addition, the weighting method (WM) is used for multiobjective optimization. The dynamic analysis is carried out using the generalized polynomial expansion method. The results show that the shaft weight and transmitted forces can be simultaneously reduced with the multilevel technique and are better than those obtained using the multiobjective optimization technique with only a single level.