The coordinate orthogonality check (CORTHOG) and multi-objective optimization considering pseudo-orthogonality as an objective function are introduced to overcome several limitations present in current model updating methods. It was observed that the use of the CORTHOG to remove four inaccurate degrees-of-freedom (DOF) was able to increase the orthogonality between mode shape vectors. The multi-objective model updating process generated a Pareto front with 38 unique optimal solutions. Four critical points were identified along the Pareto front, of which decreased the natural frequency error by greater than 2.84% and further increased the orthogonality between mode shape vectors. Therefore, it has been demonstrated that both steps of the methodology are critical to significantly reduce the overall errors of the system and to generate a finite element (FE) model that best describes physical reality. Additionally, the methodology introduced in this work generated a feasible computational runtime allowing for it to be easily adapted to widespread applications.
Multi-Objective Model Updating Optimization Considering Orthogonality
Contributed by the Design Engineering Division of ASME for publication in the JOURNAL OF COMPUTATIONAL AND NONLINEAR DYNAMICS. Manuscript received November 22, 2018; final manuscript received February 28, 2019; published online April 8, 2019. Assoc. Editor: Bogdan I. Epureanu.
- Views Icon Views
- Share Icon Share
- Search Site
Warwick, B. T., Kim, I. Y., and Mechefske, C. K. (April 8, 2019). "Multi-Objective Model Updating Optimization Considering Orthogonality." ASME. J. Comput. Nonlinear Dynam. June 2019; 14(6): 061009. https://doi.org/10.1115/1.4043086
Download citation file:
- Ris (Zotero)
- Reference Manager