Bridge fatigue reliability assessment using probability density functions of equivalent stress range based on field monitoring data

Abstract

This paper focuses on fatigue reliability assessment of steel bridges by using probability density functions of equivalent stress range based on field monitoring data. To date, existing steel bridges have experienced fatigue cracks initiated and propagated. As a result, bridge structural integrity may not be preserved safely up to its anticipated service life. For this reason, it is necessary to assess and predict bridge fatigue reliability. The AASHTO Specifications can be used to estimate capacity of structural details in the fatigue reliability assessment, whereas long-term monitoring data can be used to provide efficient information for fatigue in terms of equivalent stress range and cumulative number of stress cycles. Under uncertainties, an approach using probabilistic distributions associated with stress ranges is proposed to effectively predict equivalent stress ranges for bridge fatigue reliability assessment. The fatigue detail coefficient, A, and the equivalent stress range, S re , are both treated as random variables in the proposed fatigue reliability approach. This approach is illustrated on two existing bridges which are expected to experience finite or infinite fatigue life.