.png){kind=icon}
.png){kind=icon}
Return
Modelling of random tension processes in cable-stayed structures under random traffic loads and fatigue reliability analysis
.png){kind=icon}
PRE
AI
Z
F
X
DOI：10.1016/j.istruc.2026.112399.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Long-span cable-stayed bridges are continuously subjected to stochastic traffic loading, which may induce nonstationary and non-Gaussian cable-force responses and affect stay-cable fatigue assessment. This study proposes a stochastic-process-based framework for modelling random tension processes under traffic loading. A stochastic traffic scenario considering random arrivals, vehicle types, axle configurations, lane use, car-following, lane-changing and heavy-vehicle events is constructed. Finite-element response kernels are used to generate cable-force histories and are verified against direct finite-element moving-load calculations in a critical traffic window. The nonstationary cable-force history is represented through a finite-window increment process for local statistical modelling. Local weak stationarity, statistical regularity, candidate marginal distributions and temporal correlation models are examined using goodness-of-fit and autocorrelation metrics. An asymmetric generalized normal distribution combined with an autoregressive model (AGND-AR) is adopted to describe non-Gaussian increments and temporal dependence. The reconstructed responses are linked to rainflow counting, Goodman correction and Miner’s rule for fatigue evaluation. The basic AGND–AR(30) reconstruction gives a daily fatigue-damage error of 1.60%, which decreases to −0.19% after introducing the event-informed extreme-event correction. The framework provides a practical basis for fatigue-oriented stochastic modelling of stay cables.
Journal
IF:
4.3
Papers:
1.2W
Citations:
2.7W
Organization
Cited Papers