13th International Conference on Fracture June 16–21, 2013, Beijing, China -1- Reliability of micro/macro- fatigue crack growth behavior in the wires of cable-stayed bridge Keke Tang1,2,* 1 Key Laboratory of Pressure Systems and Safety, Ministry of Education, School of Mechanical Engineering and Power Equipment, East China University of Science and Technology, Shanghai 200237, China 2 Department of Engineering Mechanics, Southeast University, No.2 Si Pai Lou, Nanjing 210096, China * Corresponding author: kktang@ecust.edu.cn Abstract The principle of least variance is applied to evaluate the reliability of fatigue crack growth behavior in the steel wires of Runyang cable-stayed bridge. Micro/macro- crack growth model is proposed in that both micro- and macro scale effects can be included. These features are best exhibited by the normalized generic parameters that reflect the effects of material, loading and geometry. Attention is paid to the 5mm diameter wire in cables under traffic that is reflected by the additional applied tension force. Combination of theory of least variance and the dual scale model offers a new perspective for the reliability analyses of structural systems. The aim is to demonstrate there exists a time range of high reliability for the crack length history that correspond to the least variance of the time dependent R-integrals. The results coincide with the scale range established empirically by in-service health monitoring for micro-macro cracking. Such agreements are encouraging with regard to the potential application of principle of least variance in multiscale reliability assessment for multi-component and multi-function systems. Keywords Reliability, Least variance, Micro/macro- crack growth model, Cable-stayed bridge, Fatigue 1. Introduction Bridges, particularly cable-stayed bridges, are the superstar of engineering world. Aesthetically, they allow for creative freedom by varying the shape of towers and arrangement of the cables. Technically speaking, the critical components in a cable-stayed bridge are clearly the cables that behave differently and act as a structure in itself. Each of the cables consists of bundles of high strength steel wires. The structural integrity is inevitably affected by the degradation of the individual cables and wires, making replacement of cables necessary and crucial events during the life span of cable-stayed bridges. For this reason alone, rational assessment of the reliability of in-service cables in structural health monitoring can be of a big concern. However, multiple factors such as material degradation, changing environment and operating conditions, in many cases, have made structural reliability hard to predict. Fatigue crack growth behavior in wires of cables plays a substantial role in the degradation of the individual wires and cables, which stands out in the reliability analyses. Specifically for the 5mm diameter steel wires included in cables, the crack initiation and propagation are intimately related under traffic load, such that they should be treated as the same process. Fatigue damage accumulation occurs at distinctive scales. The conventional crack growth approach [1] is no longer applicable and a dual scale crack growth model in which both micro- and macro- effects are involved, should be mandated. Demonstrated in [2] are earlier researches on dual scaling. Studied in [3] was the loosening and tightening effects on the fatigue crack growth behavior of cables and wires for the cable-stayed portion of Runyang Bridge. However, emphasized here are the reliability analyses of fatigue crack growth behavior in wires. In spite of unpredictable difficulties, it is, however, desirable to let the service time span be coincident with the reliable portion. The attainment of such a goal is not straightforward because of the interactive effects of material, load and geometry. They can compensate the trade off with one another. Their delineation has been problematic. Recent works on the principle of least variance [4, 5] has shown that ‘trade-off effect’ can be made more transparent by exploring the variables including length [6-8], velocity, mass and
RkJQdWJsaXNoZXIy MjM0NDE=