Structural Safety最新文献

The traffic scenarios that may cause extreme load effects are of great importance to the safety assessment of bridge structures. The traditional simulation method of traffic flow cannot depict the distribution pattern of vehicles on the bridge deck when the maximum effect is induced. In this paper, a probabilistic Gaussian mixture model (GMM) for heavy vehicle scenarios on the bridge deck under free-flow condition is proposed for long-span bridges based on collected Weigh in Motion (WIM) data. The scenarios of extreme response under free-flow occur more frequently than congestion scenarios and are of similar value and relevance in the daily management and safety assessment of long-span bridges.

A non-stationary Poisson process is utilized to simulate the uneven occurrence of heavy vehicles in different lanes, and it is assumed that they are located within the artificially defined cells on the bridge deck. Then, Nataf transformation is employed to consider the correlation of gross vehicle weights (GVWs) within close range in the same lane. The numerical study is carried out on a long-span cable-stayed bridge to investigate the effects of correlation in GVWs and stationarity of vehicle distribution location on the structural responses. The load responses calculated by the proposed model and Monte Carlo method for different effects are compared with the values derived from code model. The results show that with the increase of the correlation level of the neighboring GVWs, the simulated responses are more prone to get extreme values, which means an increasing probability of the most unfavorable spatial distribution of on-bridge vehicles. The same results are also found under the non-stationary simulation state for vehicle location. The non-stationary Poisson process provides an efficient, highly feasible method, which is also in the safe side, for simulating the vehicle spatial distribution for specific effects.

This paper proposes an efficient solution for solving hybrid reliability problems involving random and interval variables. To meet this aim, using the soft Monte Carlo (SMC) method, a solution is proposed that breaks the random variables space into local 1-D coordinates and then, considers 1-D coordinate as an additional dimension of interval variables. Accordingly, using an optimization in increased interval variables space, the upper and lower bounds of failure probability for each 1-D problem are estimated. In addition, the total failure probabilities are presented as the mathematical expectation of the obtained probability bounds for 1-D coordinates. Then, it is shown that this approach is fit for application of univariate dimension reduction method to reduce the function calls of analysis in the optimization phase. This approach is validated by solving benchmark reliability problems as well as the application of the proposed method for solving real world engineering problems investigated by solving hybrid reliability analysis of reinforced concrete columns. It is shown that the proposed approach efficiently approximates the failure probability bound of problems with moderate nonlinear limit state functions with high accuracy.

Global reliability sensitivity analysis aims at quantifying the effects of each random source on failure probability or reliability over their whole distribution range and is highly concerned in reliability design and uncertainty control. And in practice, a structure or product usually has more than one component impacting their performance safety, which is essentially a system reliability problem. Therefore, this paper proposes a novel Bayesian-inference-based method for moment-based global sensitivity analysis of system reliability with multiple failure modes. First, the limit-state function of each component involved in the system is linearly approximated based on the reliability index. Then, the global reliability sensitivity is transformed into a problem of multivariable Gaussian probability within a given safe region where the dimension number is double of the failure modes. In this case, the Bayesian-inference-driven expectation propagation technique is introduced to solve this intractable problem in an analytical manner, based on which the closed-form solution to the global reliability sensitivity for system with multiple components is accordingly derived. Finally, a numerical case, a vehicle subjected to impact, a cantilever beam and a practical engineering application to a four-panel spaceborne deployable plane antenna are studied to demonstrate the effectiveness of the proposed method by comparison with Monte Carlo simulation.

The live load duration refers to the period when the live load is larger than a given threshold in the reference period. The smallest threshold that allows the duration to be shorter than the required length is employed as the design live load for serviceability limit states. However, the traditional method only considers the mean duration and the probability that the duration exceeds the required length is unknown. This study proposes a new algorithm to determine the probability distributions of the live load duration. A sustained or extraordinary load process is transformed into a random variable set based on the stochastic harmonic functions. Subsequently, the duration distributions can be derived by employing the load coincidence principle and probability density evolution method. Three numerical examples including one sustained load and multiple extraordinary loads are provided and the results of the proposed algorithm are compared with those of Monte Carlo simulation. The proposed algorithm allows the exact determination of design live loads based on a predefined exceeding probability. As an application, the quasi-permanent and frequent values of seven user categories are calculated when the exceeding probabilities are taken as 10%, 5% and 2%, respectively. It is found that the quasi-permanent values can increase with increasing area and the differences between the frequent and quasi-permanent values can be more than 20 times.

Under environmental action, reinforced concrete (RC) structures might suffer from reinforcement corrosion caused by the surrounding environment, dramatically reducing structural reliability and threatening social development. However, most of the existing reliability assessment methods for RC structures only focused on the structural performance at the design stage given the original unchanged environment, ignoring the effects of realistic exposure conditions and inspection results on reliability evaluation. Thus, this paper develops a general reliability assessment framework based on a Mixed Bayesian network (MBN), incorporating three modules, i.e., durability assessment, load-bearing capacity analysis, and time-dependent reliability analysis. In MBN, separate sub-BNs are built based on different modules and connected by pinch point variables where probabilistic information is transmitted via soft evidence. Besides, this framework considers time-dependent environmental parameters and two-dimensional chloride transport and their effects on reliability. Meanwhile, adjustment coefficients are applied to improve the results of the analytical mechanical model with respect to different limit states through the finite element model (FEM). The proposed MBN framework is illustrated for a corroded RC beam under a marine atmospheric environment to investigate the effects of environmental modeling, chloride transport patterns, and concrete crack inspection on reliability assessment. The results indicate that under the assumed conditions in the case study, early inspection of large cracks may significantly overestimate the failure probability by about 500%. Besides, failure probability might be underestimated by about 95%, ignoring the time-variant environment and two-dimensional chloride transport.

Reliability Analysis (RA) is a critical aspect of structural design and performance evaluation aiming to determine the probability of structural failure under given random input parameters. With modern development of modeling techniques, computational models have achieved higher fidelity but at the increased cost of computational time, which poses a significant challenge for RA. Consequently, surrogate model-assisted RA has been explored as a means of improved efficiency and accuracy. This study proposes a novel learning function, Sample-based Expected Uncertainty Reduction (SEUR), for surrogate model-assisted RA. The SEUR function uses statistical information from the metamodeling with fixed hyper-parameters to construct expected failure probability bounds to sequentially update the design of experiment (DoE). The joint probability densities of input variables are accounted for through simulation methods, including Monte Carlo (MC) and subset simulation (SS). Furthermore, the discrete simulated annealing algorithm is used to search for the optimal design point. The performance of proposed AK-SEUR function is systematically evaluated using six examples of different dimensions, failure probability levels and nonlinearities. The AK-SEUR function is demonstrated to be more effective and efficient than other popular active learning methods in dealing with nonlinear performance functions, small probabilities, and complex limit states. The proposed SEUR function has the potential to improve the efficiency and accuracy of RA, particularly in situations where computational models are time-consuming and the search for the optimal solution is challenging.

This paper addresses the problem of ranking correlated random variables according to relative importance. The importance of a variable derives from its influence on the variability of the response from a model. Applications include any input–output model for which response derivatives are available from each response analysis. Structural analysis models, i.e., finite element models, represent the specific motivation for this paper. The response derivatives are collected in a vector and transformed into a standardized parameter space. Points along that vector are transformed back to the original parameter space and utilized for the purpose of model insights and parameter ranking. Comparisons are made with the first-order Sobol sensitivity index, which requires sampling instead of the proposed single-analysis approach. Results suggest that the proposed importance measure matches the first-order Sobol index in many situations. However, for pure multiplicative “interaction” models, the first-order Sobol index tends to be anchored at the zero-correlation case. In contrast, the proposed measures are sensitive to correlation and the effect of correlation can be significant.

In seismic risk analysis, probabilistic seismic demand analysis (PSDA) is required to determine the probability distribution of structural seismic responses. One of the key steps in PSDA is to select a suite of ground motions that are consistent with seismic hazards at a target site. Current methods for selecting hazard-consistent ground motions only achieve consistency at one or several predefined periods and choose several discrete intensity levels to consider uncertainty of ground motions. To avoid drawbacks of the current methods, this study proposes a novel method for selecting hazard-consistent ground motions. In this method, a scenario earthquake set for ground motion selection is firstly obtained from seismic hazard disaggregation to a hazard level corresponding to a very small value of spectral acceleration at a specific period. Then, a suite of random target response spectra that are consistent with target site seismic hazards are simulated based on the scenario earthquake set. Finally, one suite of hazard-consistent ground motions are selected from a ground motion database. Through a numerical example, this study concludes, a suite of selected ground motions using the proposed method presents very good consistency with the target site seismic hazards. Since hazard-consistent ground motions selected using the proposed method do not depend on the building information, they can be used to accurately perform PSDA of any building and accurately predict structural seismic responses.

We propose an algorithm for selection of points from the design domain of small to moderate dimension and for failure probability estimation. The proposed active learning detects failure events and progressively refines the boundary between safe and failure domains thereby improving the failure probability estimation. The method is particularly useful when each evaluation of the performance function $g\left(x\right)$ is very expensive and the function can be characterized as either highly nonlinear, noisy, or even discrete-state (e.g., binary). In such cases, only a limited number of calls is feasible, and gradients of $g\left(x\right)$ cannot be used. The input design domain is progressively segmented by expanding and adaptively refining a mesh-like lock-free geometrical structure. The proposed triangulation-based approach effectively combines the features of simulation and approximation methods. The algorithm performs two independent tasks: (i) the estimation of probabilities through an ingenious combination of deterministic cubature rules and the application of the divergence theorem and (ii) the sequential extension of the experimental design with new points. The sequential selection of points from the design domain for future evaluation of $g\left(x\right)$ is carried out through a new decision approach, which maximizes instantaneous information gain in terms of the probability classification that corresponds to the local region. The extension may be halted at any time, e.g., when sufficiently accurate estimations are obtained. Due to the use of the exact geometric representation in the input domain, the algorithm is most effective for problems of a low dimension, not exceeding eight. The method can handle random vectors with correlated non-Gaussian marginals. When the values of the performance function are valid and credible, the estimation accuracy can be improved by employing a smooth surrogate model based on the evaluated set of points. Finally, we define new factors of global sensitivity to failure based on the entire failure surface weighted by the density of the input random vector.