Finite Element Model Update by Optimizing Weighting Factors of Static and Dynamic Responses

Abstract

This study performs a finite element model update using the natural frequency and lateral distribution factor acquired from an ambient vibration test on bridge in service. The more exact finite element model update was developed by optimizing the weighting factors of the static and dynamic responses. In this study, when updating the finite element model, the error in the lateral distribution factor is newly defined using a modal assurance criterion equation. Especially, it is important to optimize the weighting factor of static response, because static response is related to an evaluation of structure, load rating factor. The weighting factors of the lateral distribution factor and natural frequency can be optimized using a genetic algorithm which is known as finding the global solution well. The validity of the proposed method is verified by using illustrative examples whose solutions are known. Also, the load rating factor is computed when the proposed method is applied to an existing bridge. The results show that the FE model update using both static and dynamic responses yields more exact solutions than the FE model update using either static or dynamic responses alone. Also, a finite element model update with optimized weighting factors can be performed more exactly by the proposed method.