Scaling design strategy for experimental rotor systems subjected to restricted support stiffness

Abstract

It is difficult to perform practical experiments on working gas turbine rotors due to the factors of safety and implementability etc. Therefore, it is of great importance to design a well-scaled test rig that can accurately predict the dynamic characteristics of a prototype rotor. In this paper, a design method for scaled rotor systems considering dynamic similarity and restricted support stiffness is proposed. The structural similarity coefficients of the strictly scaled model (SSM) in a rotor system are obtained based on the similarity theory and scaling law. Key optimization parameters and ranges are determined through sensitivity analysis. By genetic algorithm, the geometrically distorted scaled model (GDSM) in rotor system with dynamic similarity is established to modify the SSM rotor. Moreover, modal experiments are carried out to verify the validity of the proposed scaling design method. Results show that the type, number and accuracy of optimized parameters are important in optimization design. The critical speed of the first three orders of the GDSM rotor are very close to ones in the prototype, and the maximum percentage deviation is 1.18%. Experimental results show that the rotor developed by the geometrically distorted scaling method is able to reflect the dynamic properties of prototype accurately despite the slightly distorted partial structural parameters. Such proposed design method can provide a highlight in the design and manufacturing of the key components of gas turbine rotors for laboratory.