Title: An Intelligently Designed AI for Structural Health Monitoring of a Reinforced Concrete Bridge
Author(s): William R. Locke, Stefani C. Mokalled, Omar R. Abuodeh, Laura M. Redmond, and Christopher S. McMahan
Publication: Symposium Paper
Appears on pages(s): 103-112
Keywords: artificial intelligence, Bayesian statistics, highway bridges, model uncertainty, model updating, reinforced concrete, structural health monitoring, vehicle-bridge interactions
This research employs a novel Bayesian estimation technique to perform model updating on a coupled vehicle-bridge finite element model (FEM) for the purposes of classifying damage on a reinforced concrete bridge. Unlike existing Artificial intelligence (AI) techniques, the proposed methodology makes use of an embedded FEM, thus reducing the parameter space while simultaneously guiding the Bayesian model via physics-based principles. To validate the method, bridge response data is generated from the vehicle-bridge FEM given a set of “true” parameters and the bias and standard deviation of the parameter estimates are analyzed. Additionally, the mean parameter estimates are used to solve the FEM, and the results are compared against results obtained for “true” parameter values. Furthermore, a sensitivity study is conducted to demonstrate methods for properly formulating model spaces to improve the Bayesian estimation routine. The study concludes with a discussion highlighting factors that need to be considered when using experimental data to update vehicle-bridge FEMs with the Bayesian estimation technique.