Detailed load rating analyses of bridge populations using nonlinear finite element models and artificial neural networks

Date

2013-11-01

Author

Hasançebi, Oğuzhan
Dumlupinar, Taha

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

1
views

0
downloads

For assessing load rating capacity of bridges, American Association of State Highway and Transportation Officials Manual (AASHTO) recommends a simple method, where distribution of the forces in transverse direction is estimated by axle-load distribution factors on a simply supported beam. Although the method is practical in the sense that it allows for rapid evaluation of bridge populations, it leads to over-conservative load ratings. A finite element (FE) based load rating analysis is conceived as a more accurate strategy, yet the need for constructing and analyzing a FE model for every single bridge in the population makes it impractical for load rating analyses of a bridge population. In this study an efficient method is developed for detailed load rating analyses of bridge populations through nonlinear FE models and artificial neural networks (ANNs). In this method, geometric-replica 3D FE models are used for nonlinear response analyses and load rating calculations for a sample bridge set. ANNs are then trained to learn implicit relationships between the governing bridge parameters and the resulting load ratings using this sample bridge set, and to make cost-free load rating estimations for other bridges that are not included in the set. The single-span reinforced concrete T-beam bridge population in Pennsylvania State is used to demonstrate a practical case study for application of the method. The results indicate that FE based load rating calculation procedure integrated with ANNs can be used as efficient tools for in-depth condition assessment of bridge populations.

Subject Keywords

Condition assessment, Load rating, Reinforced concrete T-beam bridge population, Artificial neural networks, Finite element modeling, Nonlinear finite element analysis

URI

https://hdl.handle.net/11511/33272

Collections

Department of Civil Engineering, Article