Highly Efficient Model Updating for Structural Condition Assessment of Large-scale Bridges

Download Final Report

CAIT project no.: CAIT-UTC-016

Fiscal Year: 2012/2013

Status: Final

Rutgers-CAIT Author(s): Guirong (Grace) Yan, Ph.D., Patrick Szary, Ph.D.

External Author(s): Charles Sikorsky, Ph.D. P.E.

Sponsor(s): California Department of Transportation, FHWA - RITA


The primary objective of this project is to propose a high-speed, highly efficient model updating technique for structural condition assessment based on the response surface (RS) method, which can be applied to large-scale real-world structures. The specific aims are listed as follows:  1) To improve the RS method by applying a more appropriate approximation function to generate a reliable surrogate model, making the process of generating the surrogate model more quickly and efficiently. A radial basis (RB) function will be a potential candidate; 2) To perform an optimization algorithm, genetic algorithm, on the obtained surrogate model to achieve the optimal physical parameters, such as the cross-sectional areas, Young’s modulus of materials and boundary conditions.