Abstract
The design of long-span cable-stayed bridge involves a large number of loads, geometric and material parameters, all of which can interact in a random fashion. It is desirable to have a total measure of the operational reliability and safety of the structural components. Based on the box-girder component strain history data of the Runyang Cable-stayed Bridge (RYCB) in China, a computer algorithm is developed to evaluate the fatigue damage that is assumed to occur in increments, according to a lognormal distribution. The corresponding probability density function is then found to obtain a fatigue reliability index β for ranking the integrity of the girders. Emphases are placed on the overall scheme of structural reliability evaluation such that the different fatigue damage criteria, probability density functions, and strain measurement techniques can be made. Finite element calculations are also used to provide strain data at locations that are not conducive for installing strain gauges, while the compatibility of measured and calculated data is made empirically. Each of the subroutine in the fatigue reliability algorithm can be altered for improvement. The flexibility allows up-dating the prediction as the monitored strains are changed by the environmental conditions. Preliminary results are first obtained to test the selected damage increments in relation to the probability function and fatigue damage criterion. Particular attention has been given to test the sensitivity of the combined governing parameters. The highly non-linear behavior of numerical calculations related to fatigue failure necessitates an in-depth understanding of the physical model. The condition under which fatigue damage accumulation is needed in contrast to the linear sum of fatigue cycles will be left for the future. Justification should be given to include the more complex issues. The aim here is to seek a simple, and yet reliable index that can account for the fatigue damage of box-girder of long cable-stayed bridges.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have