Treating Epistemic Uncertainty Using Bootstrapping Selection of Input Distribution Model for Confidence-Based Reliability Assessment

2019 ◽  
Vol 141 (3) ◽  
Author(s):  
Min-Yeong Moon ◽  
K. K. Choi ◽  
Nicholas Gaul ◽  
David Lamb
Keyword(s):  
Physical System ◽  
Epistemic Uncertainty ◽  
Bootstrap Method ◽  
Reliability Assessment ◽  
Assessment Method ◽  
Reliability Estimation ◽  
Distribution Model ◽  
Hierarchical Bayesian ◽  
Distribution Models ◽  
Input Distribution

Accurately predicting the reliability of a physical system under aleatory uncertainty requires a very large number of physical output testing. Alternatively, a simulation-based method can be used, but it would involve epistemic uncertainties due to imperfections in input distribution models, simulation models, and surrogate models, as well as a limited number of output testing due to cost. Thus, the estimated output distributions and their corresponding reliabilities would become uncertain. One way to treat epistemic uncertainty is to use a hierarchical Bayesian approach; however, this could result in an overly conservative reliability by integrating possible candidates of input distribution. In this paper, a new confidence-based reliability assessment method that reduces unnecessary conservativeness is developed. The epistemic uncertainty induced by a limited number of input data is treated by approximating an input distribution model using a bootstrap method. Two engineering examples and one mathematical example are used to demonstrate that the proposed method (1) provides less conservative reliability than the hierarchical Bayesian analysis, yet (2) predicts the reliability of a physical system that satisfies the user-specified target confidence level, and (3) shows convergence behavior of reliability estimation as numbers of input and output test data increase.

Treating Epistemic Uncertainty Using Bootstrapping Selection of Input Distribution Model for Confidence-Based Reliability Assessment

2018 ◽  
Author(s):  
Min-Yeong Moon ◽  
K. K. Choi ◽  
Nicholas Gaul ◽  
David Lamb
Keyword(s):  
Test Data ◽  
Bayesian Approach ◽  
Physical System ◽  
Epistemic Uncertainty ◽  
Reliability Assessment ◽  
Simulation Models ◽  
Distribution Model ◽  
Hierarchical Bayesian ◽  
Distribution Models ◽  
Input Distribution

To accurately predict the reliability of a physical system under aleatory (i.e., irreducible) uncertainty in system performance, a very large number of physical output test data is required. Alternatively, a simulation-based method can be used to assess reliability, but it remains a challenge as it involves epistemic (i.e., reducible) uncertainties due to imperfections in input distribution models, simulation models, and surrogate models. In practical engineering applications, only a limited number of tests are used to model input distribution. Thus, estimated input distribution models are uncertain. As a result, estimated output distributions, which are the outcomes of input distributions and biased simulation models, and the corresponding reliabilities also become uncertain. Furthermore, only a limited number of output testing is used due to its cost, which results in epistemic uncertainty. To deal with epistemic uncertainties in prediction of reliability, a confidence concept is introduced to properly assess conservative reliability by considering all epistemic uncertainties due to limited numbers of both input test data (i.e., input uncertainty) and output test data (i.e., output uncertainty), biased simulation models, and surrogate models. One way to treat epistemic uncertainties due to limited numbers of both input and output test data and biased models is to use a hierarchical Bayesian approach. However, the hierarchical Bayesian approach could result in an overly conservative reliability assessment by integrating possible candidates of input distribution using a Bayesian analysis. To tackle this issue, a new confidence-based reliability assessment method that reduces unnecessary conservativeness is developed in this paper. In the developed method, the epistemic uncertainty induced by a limited number of input data is treated by approximating an input distribution model using a bootstrap method. Two engineering examples are used to demonstrate that 1) the proposed method can predict the reliability of a physical system that satisfies the user-specified target confidence level and 2) the proposed confidence-based reliability is less conservative than the one that fully integrates possible candidates of input distribution models in the hierarchical Bayesian analysis.

A Reliability Assessment Method of Fiber Optic Gyro

2013 ◽  
Vol 760-762 ◽  
pp. 152-156
Author(s):  
Ping Huang ◽  
Shao Bin Guo
Keyword(s):  
Fiber Optic ◽  
Reliability Assessment ◽  
Temperature Stability ◽  
Assessment Method ◽  
Spectral Lines ◽  
Distribution Model ◽  
Mechanism Analysis ◽  
Fiber Optic Gyro ◽  
Failure Data ◽  
Erbium Doped

Erbium-doped fiber source for Fiber Optic Gyro (FOG) uses doped fiber to produce super fluorescence with laser pumping. It has higher output power, wide spectral lines, lower temporal coherence, good temperature stability and long life, which are perfect source for high precision FOG. To solve the problem of reliability analysis of erbium-doped fiber source for FOG with zero failure data, Weibull distribution is chosen as the life distribution model of erbium-doped fiber source on basis of the failure mechanism analysis in this paper. And Bayesian theory is used to estimate the failure rate in different time with zero failure data, then the parameters of the life model are estimated to get reliability index of erbium-doped fiber source. The method greatly decreases the number of test samples because of Bayesian estimation has take advantage of experience information, and also, it overcomes the shortcoming of relying on failure data when using traditional reliability assessment methods. So it has great value on project application.

Failure Estimation of Offshore Renewable Energy Devices Based on Hierarchical Bayesian Approach

2019 ◽  
Author(s):  
Mohammad Mahdi Abaei ◽  
Nu Rhahida Arini ◽  
Philipp R. Thies ◽  
Johanning Lars
Keyword(s):  
Renewable Energy ◽  
Reliability Assessment ◽  
Tidal Energy ◽  
Bayesian Framework ◽  
Reliability Estimation ◽  
Observation Data ◽  
Hierarchical Bayesian ◽  
Energy Devices ◽  
Mcmc Sampling ◽  
Component Failures

Abstract Improving the reliability of marine renewable energy devices such as wave and tidal energy convertors is an important task, primarily to minimize the perceived risks and reduce the associated cost for operation and maintenance. Marine systems involve a wide range of uncertainties, due to the complexity of failure mechanism of the marine components, scarcity of data, human interactions and randomness of the sea environment. The fundamental element of a probabilistic risk analysis necessarily needs to rely on operational information and observation data to quantify the performance of the system. However, in reality it is difficult to ascertain observation of the precursor data according to the number of component failures that have occurred, mainly as a result of imprecision in the failure criterion, record keeping, or experimental and physical modelling of the process. Traditional reliability estimation approaches such as Fault Tree, Event Tree and Reliability Block Diagram analysis offer simplified, rarely realistic models of this complex reliability problem. The main reason is that they all rely on accurate prior information as a perquisite for performing reliability assessment. In this paper, a hierarchical Bayesian framework is developed for modelling marine renewable component failures encountered the uncertainty. The proposed approach is capable to incorporate the conditions, which lack reliable observation data (e.g. unknown/uncertain failure rate of a component). The hierarchical Bayesian framework provides a platform for the propagation of uncertainties through the reliability assessment of the system, via Markov Chain Monte Carlo (MCMC) sampling. The advantages of using MCMC sampling has proliferated Bayesian inference for conducting risk and reliability assessment of engineering system. It is able to use hyper-priors to represent prior parameters as a subjective observations for probability estimation of the failure events and enable an updating process for quantitative reasoning of interdependence between parameters. The developed framework will be an assistive tool for a better monitoring of the operation in terms of evaluating performance of marine renewable system under the risk of failure. The paper illustrates the approach using a tidal energy convertor as a case study for estimating components failure rates and representing the uncertainties of system reliability. The paper will be of interest to reliability practitioners and researchers, as well as tidal energy technology and project developers, seeking a more accurate reliability estimation framework.

scholarly journals A Robust Prediction Model for Species Distribution Using Bagging Ensembles with Deep Neural Networks

2021 ◽  
Vol 13 (8) ◽  
pp. 1495
Author(s):  
Jehyeok Rew ◽  
Yongjang Cho ◽  
Eenjun Hwang
Keyword(s):  
Neural Networks ◽  
Species Distribution ◽  
Best Practice ◽  
Deep Neural Networks ◽  
Species Distribution Models ◽  
Species Distribution Model ◽  
Environmental Data ◽  
Distribution Model ◽  
Distribution Models ◽  
Robust Prediction

Species distribution models have been used for various purposes, such as conserving species, discovering potential habitats, and obtaining evolutionary insights by predicting species occurrence. Many statistical and machine-learning-based approaches have been proposed to construct effective species distribution models, but with limited success due to spatial biases in presences and imbalanced presence-absences. We propose a novel species distribution model to address these problems based on bootstrap aggregating (bagging) ensembles of deep neural networks (DNNs). We first generate bootstraps considering presence-absence data on spatial balance to alleviate the bias problem. Then we construct DNNs using environmental data from presence and absence locations, and finally combine these into an ensemble model using three voting methods to improve prediction accuracy. Extensive experiments verified the proposed model’s effectiveness for species in South Korea using crowdsourced observations that have spatial biases. The proposed model achieved more accurate and robust prediction results than the current best practice models.

scholarly journals Reliability estimation for drive axle of wheel loader under extreme small sample

2019 ◽  
Vol 11 (3) ◽  
pp. 168781401983684 ◽  
Author(s):  
Leilei Cao ◽  
Lulu Cao ◽  
Lei Guo ◽  
Kui Liu ◽  
Xin Ding
Keyword(s):  
Sample Size ◽  
Promising Result ◽  
Small Sample Size ◽  
Bootstrap Method ◽  
Estimation Method ◽  
Small Sample ◽  
Reliability Estimation ◽  
Fatigue Reliability ◽  
Drive Axle ◽  
Semi Empirical

It is difficult to have enough samples to implement the full-scale life test on the loader drive axle due to high cost. But the extreme small sample size can hardly meet the statistical requirements of the traditional reliability analysis methods. In this work, the method of combining virtual sample expanding with Bootstrap is proposed to evaluate the fatigue reliability of the loader drive axle with extreme small sample. First, the sample size is expanded by virtual augmentation method to meet the requirement of Bootstrap method. Then, a modified Bootstrap method is used to evaluate the fatigue reliability of the expanded sample. Finally, the feasibility and reliability of the method are verified by comparing the results with the semi-empirical estimation method. Moreover, from the practical perspective, the promising result from this study indicates that the proposed method is more efficient than the semi-empirical method. The proposed method provides a new way for the reliability evaluation of costly and complex structures.

scholarly journals A Practical Nonprobabilistic Reliability Assessment Method on Key Strata Shear Failure in Longwall Mining

2015 ◽  
Vol 2015 ◽  
pp. 1-8 ◽  
Author(s):  
Hebing Luan ◽  
Jiachen Wang ◽  
Guowei Ma ◽  
Ke Zhang
Keyword(s):  
Structural Reliability ◽  
Shear Failure ◽  
Longwall Mining ◽  
Reliability Assessment ◽  
Friction Angle ◽  
Assessment Method ◽  
Mechanical Tests ◽  
Potential Hazard ◽  
Support Design ◽  
Key Strata

Roof cutting has long been a potential hazard factor in longwall panels in some diggings in China. Meanwhile, the key strata structural reliability, which provides an assessment on the stability of overlying roof strata, may be a significant reference for support design in underground coal mines. This paper aims to investigate a practical nonprobabilistic reliability assessment method on key strata. The mechanical tests and the hollow inclusion triaxial strain tests were conducted to measure relevant mechanical parameters and in situ stress. Furthermore, against the typical failure features in Datong Diggings, China, a shear failure mechanical model of key strata is proposed. Then, an allowable-safety-factor based nonprobabilistic stability probability assessment method is given. The sensitivity of geometrical dimensions and uncertainty levels of friction angle and cohesion are further studied. It is found that thickness and span of key strata have more dominative effect on key strata’s stability compared with the other factor and the increase of uncertainty levels results in decrease of stability probability.

scholarly journals Application Natura 2000 Data For The Invasive Plants Spread Prediction

2015 ◽  
Vol 46 (4) ◽  
pp. 159-166 ◽  
Author(s):  
J. Pěknicová ◽  
D. Petrus ◽  
K. Berchová-Bímová
Keyword(s):  
Environmental Factors ◽  
Invasive Plants ◽  
Species Distribution ◽  
Natura 2000 ◽  
Distribution Model ◽  
Distribution Data ◽  
Habitat Types ◽  
Distribution Models ◽  
Occurrence Data ◽  
Heracleum Mantegazzianum

AbstractThe distribution of invasive plants depends on several environmental factors, e.g. on the distance from the vector of spreading, invaded community composition, land-use, etc. The species distribution models, a research tool for invasive plants spread prediction, involve the combination of environmental factors, occurrence data, and statistical approach. For the construction of the presented distribution model, the occurrence data on invasive plants (Solidagosp.,Fallopiasp.,Robinia pseudoaccacia,andHeracleum mantegazzianum) and Natura 2000 habitat types from the Protected Landscape Area Kokořínsko have been intersected in ArcGIS and statistically analyzed. The data analysis was focused on (1) verification of the accuracy of the Natura 2000 habitat map layer, and the accordance with the habitats occupied by invasive species and (2) identification of a suitable scale of intersection between the habitat and species distribution. Data suitability was evaluated for the construction of the model on local scale. Based on the data, the invaded habitat types were described and the optimal scale grid was evaluated. The results show the suitability of Natura 2000 habitat types for modelling, however more input data (e.g. on soil types, elevation) are needed.

Sign in / Sign up

Export Citation Format

Share Document