High-Dimensional Reliability Method Accounting for Important and Unimportant Input Variables

2021 ◽  
Author(s):  
Jianhua Yin ◽  
Xiaoping Du
Keyword(s):  
Dimension Reduction ◽  
Reliability Analysis ◽  
Limit State ◽  
Physical Models ◽  
Reliability Prediction ◽  
High Dimensional ◽  
Core Element ◽  
Reliability Method ◽  
Input Variables ◽  
Reduced Space

Abstract Reliability analysis is usually a core element in engineering design, during which reliability is predicted with physical models (limit-state functions). Reliability analysis becomes computationally expensive when the dimensionality of input random variables is high. This work develops a high dimensional reliability analysis method by a new dimension reduction strategy so that the contributions of both important and unimportant input variables are accommodated by the proposed dimension reduction method. The consideration of the contributions of unimportant input variables can certainly improve the accuracy of the reliability prediction, especially where many unimportant input variables are involved. The dimension reduction is performed with the first iteration of the first order reliability method (FORM), which identifies important and unimportant input variables. Then a higher order reliability analysis, such as the second order reliability analysis and metamodeling method, is performed in the reduced space of only important input variables. The reliability obtained in the reduced space is then integrated with the contributions of unimportant input variables, resulting in the final reliability prediction that accounts for both types of input variables. Consequently, the new reliability method is more accurate than the traditional method, which fixes unimportant input variables at their means. The accuracy is demonstrated by three examples.

High-Dimensional Reliability Method Accounting for Important and Unimportant Input Variables

2021 ◽  
pp. 1-28
Author(s):  
Jianhua Yin ◽  
Xiaoping Du
Keyword(s):  
Dimension Reduction ◽  
Reliability Analysis ◽  
Limit State ◽  
Physical Models ◽  
High Dimensional ◽  
Core Element ◽  
Reliability Method ◽  
Important Input ◽  
Input Variables ◽  
Reduced Space

Abstract Reliability analysis is usually a core element in engineering design and can be performed with physical models (limit-state functions). Reliability analysis becomes computationally expensive when the dimensionality of input random variables is high. This work develops a high dimensional reliability analysis method by a new dimension reduction strategy so that the contributions of unimportant input variables are also accommodated by the dimension reduction. The dimension reduction is performed with the first iteration of the first order reliability method (FORM), which identifies important and unimportant input variables. Then a higher order reliability analysis is performed in the reduced space of only important input variables. The reliability obtained in the reduced space is then integrated with the contributions of unimportant input variables, resulting in the final reliability prediction that accounts for both types of input variables. Consequently, the new reliability method is more accurate than the traditional method, which fixes unimportant input variables at their means. The accuracy is demonstrated by three examples.

scholarly journals Semi-Supervised Deep Learning for High-Dimensional Uncertainty Quantification

2020 ◽  
Author(s):  
Zequn Wang ◽  
Mingyang Li
Keyword(s):  
Uncertainty Quantification ◽  
Reliability Analysis ◽  
Supervised Learning ◽  
Dimensional Space ◽  
Limit State ◽  
Failure Surface ◽  
Simulation Method ◽  
High Dimensional ◽  
State Function ◽  
Latent Space

Abstract Conventional uncertainty quantification methods usually lacks the capability of dealing with high-dimensional problems due to the curse of dimensionality. This paper presents a semi-supervised learning framework for dimension reduction and reliability analysis. An autoencoder is first adopted for mapping the high-dimensional space into a low-dimensional latent space, which contains a distinguishable failure surface. Then a deep feedforward neural network (DFN) is utilized to learn the mapping relationship and reconstruct the latent space, while the Gaussian process (GP) modeling technique is used to build the surrogate model of the transformed limit state function. During the training process of the DFN, the discrepancy between the actual and reconstructed latent space is minimized through semi-supervised learning for ensuring the accuracy. Both labeled and unlabeled samples are utilized for defining the loss function of the DFN. Evolutionary algorithm is adopted to train the DFN, then the Monte Carlo simulation method is used for uncertainty quantification and reliability analysis based on the proposed framework. The effectiveness is demonstrated through a mathematical example.

Reliability analysis of wood I-joists

1993 ◽  
Vol 20 (4) ◽  
pp. 564-573 ◽  
Author(s):  
R. O. Foschi ◽  
F. Z. Yao
Keyword(s):  
Reliability Analysis ◽  
Carrying Capacity ◽  
Failure Modes ◽  
Limit State ◽  
Load Carrying Capacity ◽  
Limit States ◽  
Element Analysis ◽  
Strength Limit ◽  
Reliability Method ◽  
Load Carrying

This paper presents a reliability analysis of wood I-joists for both strength and serviceability limit states. Results are obtained from a finite element analysis coupled with a first-order reliability method. For the strength limit state of load-carrying capacity, multiple failure modes are considered, each involving the interaction of several random variables. Good agreement is achieved between the test results and the theoretical prediction of variability in load-carrying capacity. Finally, a procedure is given to obtain load-sharing adjustment factors applicable to repetitive member systems such as floors and flat roofs. Key words: reliability, limit state design, wood composites, I-joist, structural analysis.

First-Order Reliability Method for Structural Reliability Analysis in the Presence of Random and Interval Variables

2015 ◽  
Vol 1 (4) ◽  
Author(s):  
Umberto Alibrandi ◽  
C. G. Koh
Keyword(s):  
Reliability Analysis ◽  
Structural Reliability ◽  
Limit State ◽  
Computational Cost ◽  
Computational Effort ◽  
Stochastic Dynamic ◽  
First Order Reliability Method ◽  
First Order ◽  
Structural Reliability Analysis ◽  
Reliability Method

This paper presents a novel procedure based on first-order reliability method (FORM) for structural reliability analysis in the presence of random parameters and interval uncertain parameters. In the proposed formulation, the hybrid problem is reduced to standard reliability problems, where the limit state functions are defined only in terms of the random variables. Monte Carlo simulation (MCS) for hybrid reliability analysis (HRA) is presented, and it is shown that it requires a tremendous computational effort; FORM for HRA is more efficient but still demanding. The computational cost is significantly reduced through a simplified procedure, which gives good approximations of the design points, by requiring only three classical FORMs and one interval analysis (IA), developed herein through an optimization procedure. FORM for HRA and its simplified formulation achieve a much improved efficiency than MCS by several orders of magnitude, and it can thus be applied to real-world engineering problems. Representative examples of stochastic dynamic analysis and performance-based engineering are presented.

Reliability Analysis of Self-Anchored Suspension Bridge by Improved Response Surface Method

2011 ◽  
Vol 90-93 ◽  
pp. 869-873 ◽  
Author(s):  
Xiao Lin Yu ◽  
Quan Sheng Yan
Keyword(s):  
Reliability Analysis ◽  
Response Surface ◽  
Response Surface Method ◽  
Structural Reliability ◽  
Limit State ◽  
Suspension Bridge ◽  
Support Vector ◽  
Surface Method ◽  
Reliability Method ◽  
Improved Response Surface Method

The response surface method (RSM) developed in recent years is an effective way to solve the structural reliability problems with implicit performance function. In order to improve the computational efficiency and make RSM suitable well to large and complex engineering structures, the reliability analysis method based on uniform design method (UDM) and support vector machine (SVM) was proposed. UDM is adopted to select training data and SVM is used as response surface. Structural reliability index is calculated in combination with the traditional reliability analysis methods (such as, the first-order reliability method (FORM), the second-order reliability method (SORM) or Monte Carlo simulation method (MCSM)). Numerical examples show that sampled with the UDM can greatly reduce the number of samples required for training by SVM model, and a very good approximation of the limit state surface can be obtained to get the failure probability. The reliability analysis of the under serviceability limit-state of a typical self-anchored suspension bridge——Sanchaji Bridge was carried out with the improved response surface method.

scholarly journals Hybrid Reliability Analysis for Spacecraft Docking Lock with Random and Interval Uncertainty

2017 ◽  
Vol 2017 ◽  
pp. 1-9 ◽  
Author(s):  
Jianguo Zhang ◽  
Jiwei Qiu ◽  
Pidong Wang
Keyword(s):  
Reliability Analysis ◽  
Structural Reliability ◽  
Limit State ◽  
Random Variables ◽  
Simple Problem ◽  
State Function ◽  
Limit State Function ◽  
Interval Variables ◽  
Reliability Method ◽  
Hybrid Reliability

This paper presents a novel procedure based on first-order reliability method (FORM) for structural reliability analysis with hybrid variables, that is, random and interval variables. This method can significantly improve the computational efficiency for the abovementioned hybrid reliability analysis (HRA), while generally providing sufficient precision. In the proposed procedure, the hybrid problem is reduced to standard reliability problem with the polar coordinates, where an n-dimensional limit-state function is defined only in terms of two random variables. Firstly, the linear Taylor series is used to approximate the limit-state function around the design point. Subsequently, with the approximation of the n-dimensional limit-state function, the new bidimensional limit state is established by the polar coordinate transformation. And the probability density functions (PDFs) of the two variables can be obtained by the PDFs of random variables and bounds of interval variables. Then, the interval of failure probability is efficiently calculated by the integral method. At last, one simple problem with explicit expressions and one engineering application of spacecraft docking lock are employed to demonstrate the effectiveness of the proposed methods.

A Time-Variant Reliability Analysis Method Based on Stochastic Process Discretization

2014 ◽  
Vol 136 (9) ◽  
Author(s):  
C. Jiang ◽  
X. P. Huang ◽  
X. Han ◽  
D. Q. Zhang
Keyword(s):  
Stochastic Process ◽  
Reliability Analysis ◽  
System Reliability ◽  
Limit State ◽  
Complex Structure ◽  
Random Variable ◽  
State Function ◽  
Analysis Method ◽  
Reliability Problem ◽  
Reliability Method

Time-variant reliability problems caused by deterioration in material properties, dynamic load uncertainty, and other causes are widespread among practical engineering applications. This study proposes a novel time-variant reliability analysis method based on stochastic process discretization (TRPD), which provides an effective analytical tool for assessing design reliability over the whole lifecycle of a complex structure. Using time discretization, a stochastic process can be converted into random variables, thereby transforming a time-variant reliability problem into a conventional time-invariant system reliability problem. By linearizing the limit-state function with the first-order reliability method (FORM) and furthermore, introducing a new random variable, the converted system reliability problem can be efficiently solved. The TRPD avoids the calculation of outcrossing rates, which simplifies the process of solving time-variant reliability problems and produces high computational efficiency. Finally, three numerical examples are used to verify the effectiveness of this approach.

Application of Partial Safety Factors for Fitness-for-Service Assessment of Pressure Equipment With Local Metal Loss

2017 ◽  
Author(s):  
Takuyo Kaida ◽  
Shinsuke Sakai
Keyword(s):  
Reliability Analysis ◽  
Limit State ◽  
Probability Of Failure ◽  
Pressure Vessels ◽  
Metal Loss ◽  
State Function ◽  
Safety Factors ◽  
Loss Assessment ◽  
Reliability Method ◽  
Partial Safety Factors

Reliability analysis considering data uncertainties can be used to make a rational decision as to whether to run or repair a pressure equipment that contains a flaw. Especially, partial safety factors (PSF) method is one of the most useful reliability analysis procedure and considered in a Level 3 assessment of a crack-like flaw in API 579-1/ASME FFS-1:2016. High Pressure Institute of Japan (HPI) formed a committee to develop a HPI FFS standard including PSF method. To apply the PSF method effectively, the safety factors for each dominant variable should be prepared before the assessment. In this paper, PSF for metal loss assessment of typical pressure vessels are derived based on first order reliability method (FORM). First, a limit state function and stochastic properties of random variables are defined. The properties of a typical pressure vessel are based on actual data of towers in petroleum and petrochemical plants. Second, probability of failure in several cases are studied by Hasofer-Lind method. Finally, PSF’s in each target probability of failure are proposed. HPI published a new technical report, HPIS Z 109 TR:2016, that provide metal loss assessment procedures based on FORM and the proposed PSF’s described in this paper.

Effect of Dependent Interval Distribution Parameters on Reliability Prediction

2017 ◽  
Author(s):  
Yao Cheng ◽  
Xiaoping Du
Keyword(s):  
Reliability Analysis ◽  
Limit State ◽  
Distribution Parameter ◽  
State Function ◽  
Simulation And Optimization ◽  
Distribution Parameters ◽  
Reliability Methods ◽  
Input Variables ◽  
Independent Distribution ◽  
Interval Distribution

Distributions of input variables of a limit-state function are required for reliability analysis. The distribution parameters are commonly estimated using samples. If some of the samples are in the form of intervals, the estimated distribution parameters will also be given in intervals. Traditional reliability methodologies assume that interval distribution parameters are independent, but as shown in this study, the parameters are actually dependent since they are estimated from the same set of samples. This study investigates the effect of the dependence of distribution parameters on the accuracy of reliability analysis results. The major approach is numerical simulation and optimization. This study indicates that the independent distribution parameter assumption makes the estimated reliability bounds wider than the true bounds due to interval samples. The reason is that the actual combination of the distribution parameters may not include the entire box-type domain assumed by the independent interval parameter assumption. The results of this study not only reveal the cause of the inaccuracy of the independent distribution parameter assumption, but also demonstrate a need of developing new reliability methods to accommodate dependent distribution parameters.

Sign in / Sign up

Export Citation Format

Share Document