جستجوی مقالات مرتبط با کلیدواژه « failure probability » در نشریات گروه « عمران »

The Hasofer-Lind and Rackwitz-Fiessler (HLRF) algorithm, which is based on the first-order reliability method (FORM), is widely used to estimate failure probability, reliability index, and design point in structural reliability analysis. However, due to the high nonlinearity of the limit state surface, the HLRF algorithm can be unstable. To address this issue, this paper proposes an optimization method to locate and estimate the design point in the standard normal space and calculate the corresponding failure probability. The reliability problem is solved using sequential least squares programming (SLSQP) to improve accuracy, robustness, and efficiency. SLSQP replaces the quadratic programming problem with a linear least-squares problem, using a stable LDL factorization of the Hessian of the Lagrangian equation. The initial optimization problem is converted into a minimum distance optimization problem with a lower bound constraint. To eliminate linearization errors, the probability expectation method with rotation directions space is employed. The proposed algorithm is demonstrated in several benchmark numerical examples with both explicit and implicit limit state functions. Its fast convergence rate is a notable feature of the proposed algorithm, which enhances its competitiveness in structural reliability analysis.

Bridges are a critical part of the urban and suburban transportation network, so they are supposed to be designed to sustain earthquake induced damages to be utilized after earthquake. Various parameters can affect the behavior and probability of failure of a bridge and the present work aims to evaluate the effects of changes in structural parameters on the probability of failure in isolated concrete bridges OpenSees software is used for simulating and analyzing 16 different bridge models. Incremental dynamic analysis is conducted using this software and IDA and fragility curves of models are derived and presented. The results show that the probability of failure decreases with the increase of the pier diameter, concrete compressive strength, yield strength of longitudinal rebars and diameter of longitudinal bars. Also increasing the stiffness of elastic isolator and decreasing confined diameter of pier results in increasing the probability of failure. Furthermore, results show that, the probability of failure is more sensitive on the variation of pier confined diameter, yield strength of longitudinal rebars, diameter of longitudinal bars and the stiffness of elastic isolators in comparison with the variation of concrete compressive strength.

By studying structural analyses in the space of uncertainty, it is apparent that there has been a little research on the effects of the type of earthquake on probabilistic analysis of the structure with various random distributions and coefficients of variation. Therefore, this research attends to this shortcoming. In order to do so, a one-story steel frame has been used. Structural properties such as density, elastic modulus and beam and column dimensions are considered as uncertain parameters with each having three uniform, normal and log-normal types of density function and three coefficients of variation of 5%, 10% and 15%. This structure excited by two far-field earthquakes (El Centro and Hachinohe), and two near-field ground accelerations (Northridge and Kobe). By applying Monte Carlo simulation method in this research, desired random variables are generated and applied in probabilistic analysis.

Engineering structures, during construction and operation, are under the influence of uncertainties including: dimensional parameters, materials specification, and loading. Among engineering structures, concrete structures are more sensitive to these uncertainties with respect to performance problems, environmental conditions and diversity of the components. In this study, the probability performance of roller compacted concrete (RCC), used in pavements, was investigated with different amounts of Lumachelle and pozzolan. The water to cement ratio and the amount of pozzolan and Lumachelle fine aggregates were considered as random variables and also the dimension reduction method (DRM), in combination with Monte Carlo simulation, were used for reliability and sensitivity analysis of this kind of concrete. The experimental design was based on the DRM, and compressive strength and water adsorption tests were performed on the specimens. Results showed that the probability of failure due to water adsorption is 0.17, which is more than the probability failure of compressive strength (0.021) in the RCC. Also, the probability of failure increased to 0.24 by assuming the concrete as a system.

Connections in steel moment frames are idealized in full pinned and full rigid conditions. Because with this assumption, in spite of real behavior of connection, real story drifts are less anticipated and maybe frame is designed without performance of bracing. There are several methods for modeling actual behavior of semi rigid connections. In this method a connection with certain rigidity is modeled by a rotational spring with corresponding stiffness. This stiffness is achieved by certain formula. In other words, each percent of rigidity corresponds to one rotational spring stiffness. In this research in order to evaluate the real behavior of connection in analysis and designing process and fracture probability one frame including four stories and one bay with three types of connection has been modeled and designed in ETABS. Each model has an individual rigidity which is equal to 10, 75 and 90 percent. With respect to maximum drift and different PGA in roof, probabilities of low, medium, high and complete fracture were calculated. For this purpose, with applying different PGA to modeled frames, amounts of drift in the roof are achieved. Then these values are compared with given values in American code. Finally, investigation showed that when rigidity in frame connections increases, the probability of frame fracture decreases. In other words, fully rigid assumption of connection in analysis process leads to decreasing in real probability of fracture in frames which is a noticeable risk in building designing processes.

The failure probability of structures are rather small and therefore calculation of structural reliability generally has a high computational cost. In order to reduce computational costs, this articles proposes a hybrid approach based on combination of the least squares support vector regression and two advanced Monte Carlo importance sampling and Latin hypercube sampling. Two frames and one truss example are used to evaluate the performance of the proposed algorithm. Results demonstrate that proposed method provides an accurate estimation of failure probability and that the computational costs are lower than those of other methods

The inherent uncertainty in the behavior of a structure can lead to significant discrepancies between the results of the analytical model and the actual behavior of the structure. The probabilistic characteristics of the material, errors in construction, and the uncertainty of seismic loads could account for major causes of discrepancy between analytical results and the actual response of the structure. In this regard, inevitable uncertainties in the system’s behavior should be recognized, and the probability of the structure’s failure should be calculated for the desired level of performance. The availability of sophisticated computational tools provides the engineering community with the opportunity to estimate the performance of structures by computing the probability of occurrence for a specific state in the structure. Simulation techniques, such as Monte Carlo simulation, are efficient and promising computational methods in the context of reliability analyses to obtain the required knowledge about the probabilistic behavior of structural systems. In this paper, the effect of the tuned mass damper (TMD) on the performance of steel moment-resisting frames has been studied using the Monte Carlo simulation technique. The failure probability for the stories of the structure has been calculated based on the relative displacement of each story. The capacity and stiffness of the structural elements have been modeled through probabilistic fiber-discretized sections for each element. Also, the probabilistic effect of seismic loading has been taken into account by generating records with fully stochastic characteristics in the time and frequency domain. The uncertainty in the damping characteristics and seismic mass of the structure is also considered. The results indicate that for far-field earthquakes with low mean period, TMD could reduce the failure probability of the structure, with different effects on each story. The maximum density of relative displacement has been shifted towards small values, mainly for higher stories, by using the TMD on the roof story of the structure

Abstract Accurate calculation of reliability index is very important for the reliability analysis of structures. In some limit state functions with nonlinear characteristic and several local optimum design points, the computational reliability methods may not appropriately determine the failure probability, and iterative reliability approaches may be converged to local optimum design points. In this paper, formulation of nonlinear conjugate gradient methods such as: Fletcher & Reeves (FR) and Dai & Yuan (DY) were proposed for computing the reliability index. Then, the efficiency, capacity and robustness of the conjugate gradient were investigated through several reliability problems. The FR and DY algorithms were diverged in some examples. Therefore, a new modified nonlinear conjugate gradient method was proposed by using the hybrid optimization algorithm to ensure accurate convergence of reliability index. The results illustrated that the modified nonlinear conjugate gradient methods were converged with less run-times and number of iterations in comparison with the FR and DY methods, and robust, efficient and stable algorithms are to calculate the structural failure probability.