Probabilistic analysis-based vibration characteristics of intentionally mistuned blisk

Abstract

The machining error, wear and material dispersion lead to serious vibration and large local instability in aeroengine’s blisk. To restrain the vibration and reduce the fluctuation, the intentional mistuning is adopted, and a methodology named multi-stage multi-objective coordinated extremum response surface method-based reduced-order model (MSMO-CERSM-ROM) is proposed to investigate the probabilistic distribution of intentionally mistuned blisk. The basic idea of this approach is that the large complex structure involving multiple disciplines is divided into substructures only including single discipline. This method not only can dramatically improve the computational efficiency and accuracy of the multi-objective and multi-stage response surface model but also effectively solve the nonlinear and implicit function. The probabilistic analysis of intentionally mistuned blisk via MSMO-CERSM is investigated based on the deterministic analysis via the ROM. Finally the intentionally mistuned blisk is solved compared with the tuned and passive mistuned blisk to verify the validity and rationality of the present methodology. These results manifest that the stability is strengthened, and the output response are reduced, which plays an important role in probabilistic design for complex mechanical structures