Fatigue lifespan assessment of stay cables by a refined joint probability density model of wind speed and direction

Stay cables are crucial load-bearing components of long-span cable-stayed bridges. Because they are sensitive to wind, the wind load is an important control load in their safety and fatigue evaluation. To consider the effects of variations in the wind speed and direction, a joint probability density model of wind speed and direction is proposed for evaluating the fatigue lifespan of a stay cable. For this purpose, an angular-linear model is introduced to coordinate the marginal distribution of wind speed and direction. The marginal distribution of wind speed comprises a Gumbel model, while that of the direction comprises a refined mixture of a von Mises distribution model. Subsequently, based on the dynamic simulation of a coupled wind-vehicle-bridge system, an assessment method for fatigue lifespan is developed by considering the variation of traffic flow with four load levels and variations in wind speed and direction. The results indicate that the refined angular-linear distribution model yields considerable fitting results for the measured data with respect to the wind speed and direction. Although the response and fatigue damage of a stay cable mainly depend on its length and location, the effect of the direction variation of the wind load is also not negligible. Additionally, the results from the assessment method show that the fatigue lifespan of a stay cable on the windward side of the local prevailing wind direction is relatively shorter than that of the leeward side.