Search

Structural resistance deterioration is by nature a non-increasing stochastic process with autocorrelation on the temporal scale. The Gamma process is often used to describe the stochastic behavior of resistance deterioration. With Gamma-based deterioration models, the calculation of time-dependent reliability presents a serious challenge, and often Monte Carlo simulation is the only solution to this problem. This paper derives a closed-form solution for time-dependent reliability of aging structures, in which the resistance deterioration is described by a Gamma process and the applied load is modelled as a Gaussian process with a constant standard deviation. The accuracy of the proposed method is verified through a comparison with Monte Carlo simulation results, and its applicability is further illustrated in a time-dependent reliability analysis of an existing highway bridge whose vehicle load is modelled using weigh-in-motion (WIM) data. It is found that the measured vehicle load has a relatively small uncertainty, and the uncertainty associated with resistance deterioration is crucial to the reliability assessment because it dominates the overall uncertainty in the reliability calculation. [ABSTRACT FROM AUTHOR]

This study addresses the ultimate structural and fatigue damage loads of a spar-type offshore floating wind turbine under joint excitations of wind and wave. Aero-hydro-servo-elastic coupled analysis is performed in time-domain to capture the dynamic responses of the floating wind turbine. Based on the mean up-crossing rate method, the short-term ultimate structural load is estimated. The cumulative fatigue damage load is computed with the S-N method. It is shown that the low-level ultimate load is mostly influenced by wind forces whereas the high-level ultimate load is more closely related to wave forces. The wave excitations dominate the fatigue damage at tower top and tower base, whereas the mooring line fatigue damage is more sensitive to the wind forces. [ABSTRACT FROM AUTHOR]

Extreme value predictions of the roll motion of ships are considered within the framework of the First Order Reliability Method focusing on the applicability of the Poisson assumption for the peak value statistics. The examples considered are a ship rolling in dead sea condition under the combined action of waves and wind and a ship experiences parametric rolling. • Extreme value predictions of ship roll motion are considered within the framework of the First Order Reliability Method. • The focus is on the applicability of the Poisson assumption for the peak value statistics for nonlinear responses. • A numerical investigation is conducted to check the feasibility of applying the Vanmarcke's factor to the present non-linear responses. • In addition to FORM, the Second Order Reliability Method (SORM) is applied in the calculation of the reliability index. • Two different cases of large roll motions of ships are considered. One is rolling in beam seas, and the other is parametric rolling in longitudinal seas. [ABSTRACT FROM AUTHOR]

The main focus of this paper is the development of an analytical moment-based procedure for calculating first passage times of a stationary non-Gaussian structural response. In the procedure, Winterstein’s polynomials based on response moments (skewness, kurtosis, etc.) are used to map a non-Gaussian structural response into a standard Gaussian process, and then the mean up-crossing rate of the original response process is estimated from the mapped standard Gaussian process. Through introducing the mean clump size and the probability of instantaneous failure of the original response process, the procedure accounts for effects of clumping and initial conditions on the up-crossing rate, and finally, first passage time. An analysis of a linear single-degree-of-freedom system (SDOF) system excited by a quadratic forcing function demonstrates the usage of the procedure. Comparisons between the results from the procedure and those from the Monte Carlo simulation method are also performed in the example. [Copyright &y& Elsevier]