Statistical Analysis of Online Strain Response and Its Application in Fatigue Assessment of a Long-Span Steel Bridge

ZX Li,THT Chan,R Zheng
DOI: https://doi.org/10.1016/s0141-0296(03)00174-3
2003-01-01
Abstract:This paper covers reliability assessment of the fatigue life of a bridge-deck section based on the statistical analysis of the strain–time histories measured by the structural health monitoring system permanently installed on the long-span steel bridge under study. Through statistical analysis of online strain responses in the frequency domain using multiple linear regression, a representative block of daily cycles of strain history is obtained. It is further assumed that all cycles of online strain response during bridge service are repetitions of the representative block. The rain-flow counting method is then used to determine the stress spectrum of the representative block of daily cycles. The primary assessment of fatigue life at a given value of failure probability is undertaken for the sample component in a bridge-deck section by using the classification of details for welded bridge components and the associated statistical fatigue model provided by the British Standard BS5400. In order to evaluate bridge fatigue at any value of failure probability, a modified probability model is proposed based on BS5400. The fatigue life of the considered component in the bridge-deck section is then evaluated for some other values of probability of failure which are not included in BS5400 by use of the modified probability model. The analytical results show that the modified model is practical for reliable evaluation of the service life of existing bridges under random traffic loading.
What problem does this paper attempt to address?