Collaborative Research: Fatigue Damage Prognosis for Slender Coastal Bridges

Project: Research project

Project Details


Transportation network, including highways and transit, are vital to the nations economy. With many slender coastal bridges built to cross rivers, harbors or bays, the safety of these bridges has been much more important when the coastal regions have had 40% U.S. population. As many catastrophic disasters originate from failures in local details, effective decision making, especially when damages or cracks are detected in early age during the life-cycle, are extremely import. Structural evaluation and risk assessment in structural component and system level should be carried out by defining limit states in different levels with acceptable risk to personnel, environment, and economy.
In this project, we propose maximum entropy based Bayesian network (MEBN) to predict multi-scale damage propagation for decision making. Three levels of nodes, namely, the material level, structural level, and system level, are built for slender coastal bridges in the Bayesian network. The inter-correlations between different levels of nodes are built by using coupled dynamic analysis and corrosion fatigue damage analysis. With the propagation of local cracks and progressive failure modes, the load path discontinuity and progressive failure modes will be detected and the MEBN will be updated. The concurrent analysis approach will be carried out for corrosion fatigue damage analysis and local stress analysis to build additional correlation between nodes in material level and structural level. Uncertainties and conditional probabilities will be propagated through the proposed Bayesian network and the system level reliability will be reassessed. Specifically, we aim at corrosion fatigue damage analysis, coupled structural dynamic analysis, and developing MEBN, which includes the following research activities: (1) to explore the fundamental corrosion fatigue damage progress in steel and steel components by integrating multi-resolution experimental testing and coupled multi-level damage analysis; (2) to conduct correlated non-stationary wind and wave simulation and coupled structural dynamic analysis with consideration of additional structural failure modes and load path discontinuities; (3) to investigate a new entropy-based Bayesian network for information fusion in the multi-scale damage prediction and risk assessment of slender coastal bridges.
Effective start/end date9/1/158/31/20


  • National Science Foundation (NSF): $235,000.00

Fingerprint Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.