Publication Date
2010
Document Type
Thesis
Committee Members
Jeffrey M. Brown (Committee Member), Ravi C. Penmetsa (Committee Member), Joseph C. Slater (Advisor)
Degree Name
Master of Science in Engineering (MSEgr)
Abstract
Mistuning prediction in integrally bladed rotors is often done with reduced order models that minimize computational expenses. A common model reduction technique used for mistuning applications is the component mode synthesis method. In this work, two modern component mode synthesis methods are used to generate mistuned response distributions that will be used to determine if the two methods are statistically indistinguishable. The first method, nominal mode approximation, assumes an airfoil geometric perturbation alters only the corresponding substructure modal stiffnesses while its mode shapes remain unaffected. The mistuned response is then predicted by a summation of the nominal modes. The second method, non-nominal mode approximation, makes no simplifying assumptions of the dynamic response due to airfoil geometric perturbations, but requires recalculation of substructure matrices and mode shapes with each iteration. The number of retained fixed interface normal modes for the non-nominal method are increased until there is satisfactory accuracy compared to full finite element model results. Each approach is employed for calculating the mistuned response of a simple academic rotor and an advanced rotor with complex geometries. Three different veering regions are investigated in the advanced test case. Results indicate there is minimal difference between response distributions generated by the nominal and non-nominal methods for the academic rotor. Large differences were observed for the advanced rotor, where the nominal method typically predicted conservative response levels larger than non-nominal predictions.
Page Count
180
Department or Program
Department of Mechanical and Materials Engineering
Year Degree Awarded
2010
Copyright
Copyright 2010, all rights reserved. This open access ETD is published by Wright State University and OhioLINK.