Adaptive Surrogate Modeling for Time-Dependent Multidisciplinary Reliability Analysis

Hu, Zhen; Mahadevan, Sankaran

doi:10.1115/1.4038333

Cited by 18 publications

(7 citation statements)

References 31 publications

(62 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such a sensitivity analysis will benefit the enrichment step as it will allow to enrich only the disciplinary GP responsible for most of the variation of the objective minimum as in the current EGMDO approach all the disciplinary GPs are enriched during the iterative procedure. It should be noted that this idea has been developed by [12] in the context of reliability analysis. Finally, constraints handling should also be considered in the future.…”

Section: Discussionmentioning

confidence: 99%

“…This test case is inspired by the one used in [12] and [7] and deals with a compound cylinder. Figure 24 presents the compound cylinder as well as the the two sub systems, the inner and outer cylinder respectively.…”

Section: Presentation Of the Test Casementioning

confidence: 99%

“…Such an approach has great potential as it allows to uncouple the MDO problem (as the IDF method) and to focus the computation of the costly disciplinary solvers only on interesting areas with respect to the optimization. It could be noted that the idea of disciplinary Gaussian surrogate models constructed in an adaptive way with respect to a given improvement criterion has been recently proposed in [12] in the context of time dependent reliability analysis of a multidisciplinary system, the numerical procedure to deal with the surrogate model uncertainty propagation and the enrichment criterion being however different from the one proposed in the present article. One can also note that, in the context of disciplinary solvers directly chained (no feedback coupling), the use of disciplinary Gaussian processes has been proposed in [18] and [26] for the adaptive construction of a global surrogate model (in the particular case of direct coupling of Gaussian surrogate models this is nevertheless simplified as analytical expressions of mean and variance of the final quantity of interest can be derived analytically).…”

mentioning

confidence: 93%

See 2 more Smart Citations

Towards an efficient global multidisciplinary design optimization algorithm

Dubreuil

Bartoli

Gogu

et al. 2020

Struct Multidisc Optim

View full text Add to dashboard Cite

This article proposes a new surrogate based Multidisciplinary Design Optimization algorithm.The main idea is to replace each disciplinary solver involved in a non linear Multidisciplinary Analysis by Gaussian Process surrogate models. Although very natural, this approach creates difficulties as the non linearity of the Multidisciplinary Analysis leads to a non Gaussian model of the objective function. However, in order to follow the path of classical Bayesian optimization such as the Efficient Global Optimization algorithm, a dedicated model of the non Gaussian random objective function is proposed. Then, an expected improvement criterion is proposed to enrich the disciplinary Gaussian processes in an iterative procedure, that we call Efficient Global Multidisciplinary Design Optimization (EGMDO). Such an adaptive approach allows to focus the computational budget on areas of the design space relevant only with respect to the optimization problem. The obtained reduction of the number of solvers evaluations is illustrated on a classical MDO test case and on an engineering test case.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Presentation Of the Test Casementioning

confidence: 99%

mentioning

confidence: 93%

See 1 more Smart Citation

Towards an efficient global multidisciplinary design optimization algorithm

Dubreuil

Bartoli

Gogu

et al. 2020

Struct Multidisc Optim

View full text Add to dashboard Cite

show abstract

“…The analytical computation of the failure probability is usually difficult and that can be done by approximation methods (Xu and Kong, 2018). Defining systems momentary behavior under time varying inputs accompanied by cluster of coupling variables possesses significant computational challenges in reliability analysis (Hu and Mahadevan, 2017). Eryilmaz (2018) declared that the reliability of the system defined the probability that the performance of the system will be at least to that given level.…”

Section: Methodological Guidelinesmentioning

confidence: 99%

Investigating economic replacement policy under uncertainty for managerial application based on grey-reliability approach using QCC

Prabhu

Sahu

2022

View full text Add to dashboard Cite

PurposeThe main purpose of present study is to model the replacement policy under uncertainty for managerial application based on grey-reliability approach by considering the subjective views of quality control circle (QCC). The study objectively links the optimality between individual replacement and group replacement policies for determining the minimum operational costs. The integrated framework between QCC, replacement theory, grey set theory and supply chain management is presented to plan replacement actions under uncertainty.Design/methodology/approachThe study proposes the concept of grey-reliability index and built a decision support model, which can deal with the imprecise information for determining the minimum operational costs to plan subsequent maintenance efforts.FindingsThe findings of the study establish the synergy between individual replacement and group replacement policies. The computations related to the numbers of failures, operational costs, reliability index and failure probabilities are presented under developed framework. An integrated framework to facilitate the managers in deciding the replacement policy based on operational time towards concerning replacement of assets that do not deteriorate, but fails suddenly over time is presented. The conceptual model is explained with a numerical procedure to illustrate the significance of the proposed approach.Originality/valueA conceptual model under the framework of such items, whose failures cannot be corrected by repair actions, but can only be set by replacement is presented. The study provides an important knowledge based decision support framework for crafting a replacement model using grey set theory. The study captured subjective information to build decision model in the ambit of replacement.

show abstract

“…[36] is still effective when stochastic processes are involved. Adaptive Kriging-based methods have drawn much attention from the literature [37,38]. Hu and Mahadevan [39] modified the learning function U which was initially proposed for timeindependent reliability analysis to adapt time-variant reliability analysis [40] and developed a single-loop Kriging surrogate modeling method.…”

Section: Introductionmentioning

confidence: 99%

Importance Sampling for Time-Variant Reliability Analysis

2021

View full text Add to dashboard Cite

Importance sampling methods are extensively used in time-independent reliability analysis. However, the kind of methods is barely studied in the field of time-variant reliability analysis. This paper presents an importance sampling method for time-variant reliability analysis. It increases the probability of sampling failure trajectories of a time-variant performance function. To develop the method, the instantaneous performance function at a predefined time instant is regarded as a time-independent one. A time-independent importance sampling is first implemented on the instantaneous performance function in order to obtain instantaneous samples of stochastic processes and random variables. Then, conditional trajectories of stochastic processes are generated on the condition of instantaneous samples achieved above, which utilizes the correlationship among instantaneous uncertainties at different time instants associated with stochastic processes. Subsequently, trajectories of the time-variant performance function are obtained. Validation results show that comparing with crude Monte Carlo simulation, the proposed method remarkably increases the probability of sampling failure trajectories. The efficiency and accuracy of the proposed method are demonstrated.

show abstract

Adaptive Surrogate Modeling for Time-Dependent Multidisciplinary Reliability Analysis

Cited by 18 publications

References 31 publications

Towards an efficient global multidisciplinary design optimization algorithm

Towards an efficient global multidisciplinary design optimization algorithm

Investigating economic replacement policy under uncertainty for managerial application based on grey-reliability approach using QCC

Importance Sampling for Time-Variant Reliability Analysis

Contact Info

Product

Resources

About