Search

The problem of assessing the damage to and remaining life of structures subjected to fatigue is explored, with particular emphasis on railroad bridges, highway bridges, and offshore structures. In railroad structures, the fatigue reliability estimates are typically calculated based on fatigue life predictions. The resulting predictions are used for budgeting purposes rather than for scheduling repairs. Examples are given of economic decision based on fatigue life calculations. Fatigue reliability estimates for highway bridges are also typically based on fatigue life calculations. Target reliability values have been suggested and are used in both the design of new bridges and the evaluation of existing bridges. Given the economics of offshore structures, it is feasible to develop more elaborate fracture mechanics models. These models have the advantage of explicitly allowing updated estimates of fatigue reliability based on inspection results. The economics of inspection of offshore structures also dictate...

The need for reassessment of the fatigue life of existing structures is increasing as the world’s infrastructure ages. A fatigue life reassessment typically begins with an assessment of the current condition of the structure. The condition assessment techniques range from visual inspection to X-ray inspection or detection of acoustic emissions. The fatigue reliability of the structure can be estimated from probabilistic fatigue life or fracture mechanics models. The data obtained from the condition assessment can be combined with these models to estimate the remaining service life of a structure using Bayes’ theorem. Simulation techniques are often used to facilitate these calculations. If the remaining service life is inadequate, it may be desirable to repair the structure; however, repairs must be performed carefully to provide the desired benefit. On the other hand, economic factors may dictate a course of action other than repair, such as replacing the structure or changing the operation of the structure.

This study is intended to provide experimental verification of the validity of analytical methods for the prediction of the response to loads of flexibly connected steel frames. To this end, test data are provided for the behavior of one particular connection type. These data are formulated for incorporation in a displacement matrix analysis, which is then used to predict the response to load of flexibly connected, statically indeterminate steel frames. This predicted behavior is then compared to measured moments, joint rotations, and deflections of loaded frames. It is concluded on the basis of these limited test data that the method is well able to predict frame response to load cycles, that linearly elastic analysis is sufficient to predict frame behavior under service conditions, and that these frames tend to shake down to their elastic state under high load cycles.