Reliability analysis of wind-sensitive structures

Abstract

SummaryThe theoretical background of practical structural reliability methods encompassing numerical integration, approximate methods and simulation techniques is reviewed in the context of windexcited structures. Their relative merits, shortcomings and limitations are addressed from the standpoint of their accuracy and computational efficiency. Quantitative reliability estimates of a concrete chimney are made, using different analysis procedures. Multiple potential failure modes are represented by the exceeding of the ultimate moment capacity at any level of the chimney height. The general bounds on the system failure probability are expressed in terms of the failure probability of the individual modes. The narrower bounds are established based on the existing theory by taking into account not only the failure probability of the individual modes, but also the joint failure probabilities in any two modes. Based on the findings of this study, it is suggested that the advanced first-order second-moment approximation and the simulation methods, which combine the Monte Carlo technique with variance reduction techniques, may provide accurate results for practical reliability analysis of wind-excited structures.

An efficient framework for the elasto-plastic reliability assessment of uncertain wind excited systems

“…It is also the basis on which state-of-theart performance-based design is founded [5,6]. Recently there has been significant interest in developing specific reliabilitybased procedures for assessing the performance of wind excited structures [7][8][9][10][11][12][13][14][15][16][17]. However, as in classic reliability analysis, the limit states indicating structural failure are associated with first yield of the structural system.…”

Reliability analysis of wind-excited structures

“…Failure probability is accomplished assuming capacity and demand (or response and discomfort threshold) as normal or lognormal distributed. Based on this procedure, Rodjani and Wen (1981), Schuëller et al (1983) and Kareem (1988Kareem ( , 1990) studied serviceability and ultimate limit states considering a large number of parameters both of the load and the structure, also taking into account different failure modes. Results indicate the significant contribution of the uncertainties related to the dynamic structural properties and wind in the overall reliability.…”

“…Moreover, statistical information of the model parameter distribution is usually quite scarce, hardly arriving at the first or second statistical moments. Therefore, numerical evaluations of the uncertainties connected with the structural response to wind action are carried out making recourse to numerical simulations, approximate procedures, or both (see Kareem, 1990 or Solari, 1997 for a general frame of methods).…”

Computation of failure probability via hierarchical clustering

“…In the case of sampling-based methods, one way of reducing their computational effort, especially in the case of small failure probabilities, is given by the use of variance reduction strategies with advanced sampling schemes, which have the aim of limiting the number of simulations without losing accuracy. These methods can be categorized as importance sampling-based methods (Rubinstein and Kroese, 2011;Robert and Casella, 2004), splitting and subset simulation (SS) methods (Ching et al, 2005;Au and Beck, 2001), and methods that focus on improving the quality of the samples, such as quasi-Monte Carlo (Fang and Wang, 1994;Nie and Ellingwood, 2004;Li and Chen, 2008), common random numbers (Hammersley and Handscomb, 1964), antithetic variates (Hammersley and Morton, 1956;Kareem, 1990).…”