Reliability calculation of time-consuming problems using a small-sample artificial neural network-based response surface method

Lehký, David; źOmodíková, Martina

doi:10.1007/s00521-016-2485-3

Cited by 34 publications

(19 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The general principle of surrogate modeling (also known as metamodeling) is to replace the original model, which requires enormous computational effort, with an approximated (simpler) model whose evaluation is not so time‐consuming. Metamodels that have gained popularity among researchers over the last few decades include an ANN, polynomial chaos, support vector machine, and Kriging metamodel . In case of the practical applications presented in the following section ANN‐ based surrogate models were used.…”

Section: Surrogate Modelsmentioning

confidence: 99%

“…For more details on ANN‐based surrogate modeling see Lehký and Šomodíková . Let us note that two independent types of ANN modeling were performed in the case of NNE‐based sensitivity analyses in both case studies: one was employed in surrogate modeling (utilized in all sensitivity methods) and the second one was used directly in NNE‐based sensitivity analysis.…”

Section: Surrogate Modelsmentioning

confidence: 99%

See 1 more Smart Citation

A comparison of sensitivity analyses for selected prestressed concrete structures

Lehký

Pan

Novák

et al. 2018

Structural Concrete

Self Cite

View full text Add to dashboard Cite

Three sensitivity analysis methods are employed to achieve the optimum selection of the dominant random variables of selected concrete structures. The first of these methods uses the nonparametric rank-order statistical correlation between the basic random input variables and the structural response output variable. The second is neural network ensemble-based sensitivity analysis and the last of them is sensitivity analysis in terms of coefficient of variation. All of the methods were utilized and compared for two selected concrete structures: a prestressed concrete bridge made of MPD girders, and T-shaped prestressed concrete roof girder. The obtained information was used to set up a stochastic model and response surfaces in an optimum manner and was employed in the subsequent determination of selected uncertain design parameters followed by load-bearing capacity and reliability assessment.artificial neural network, prestressed concrete, sensitivity analysis, structural reliability, surrogate modeling

show abstract

Section: Surrogate Modelsmentioning

confidence: 99%

Section: Surrogate Modelsmentioning

confidence: 99%

A comparison of sensitivity analyses for selected prestressed concrete structures

Lehký

Pan

Novák

et al. 2018

Structural Concrete

Self Cite

View full text Add to dashboard Cite

show abstract

“…Für die vorliegende Problemstellung bietet sich die „Adapted Latin Hyper Cube Sampling”‐Methode (HSLHS) unter Zuhilfenahme der FReET Software an. Die „Latin Hyper Cube Sampling”‐Methode ermöglicht eine Reduktion der sehr hohen Simulationsanzahl der Monte‐Carlo‐Methode auf wenige hundert Simulationen . Mit dieser Methode ist aufgrund der folgenden Strategie eine weitere wesentliche Reduktion möglich: Aus den Verteilungsfunktionen der streuenden Modelleingangsgrößen wird nach dem LHS‐Grundprinzip ein umfangreiches Set an Stichproben für die nichtlinearen Modellbildungen geschaffen.…”

Section: Analyse Des Stabilitätsversagensunclassified

Zuverlässigkeit schlanker Betonstützen

Strauß¹,

Benko

Tãubling³

et al. 2017

Beton und Stahlbetonbau

View full text Add to dashboard Cite

In diesem Beitrag werden Berechnungen und vergleichende experimentelle Ergebnisse des Stabilitätsverlusts von schlanken Stahlbetonstützen aus einer Versuchsreihe präsentiert. Das primäre Ziel der Arbeit war der Vergleich der Bemessungsmethoden bei der Ermittlung der globalen Zuverlässigkeit schlanker Stahlbetonstützen. Der Schlankheitsgrad der Stützen betrug λ > 90, wobei die untersuchten Stützen nach EN 1992‐1‐1 dimensioniert wurden. Die oben erwähnten experimentellen Untersuchungen dienten zudem als Grundlage für die Bewertung und Kalibrierung der deterministischen und probabilistischen nichtlinearen Modellbildung. Diese probabilistische Modellbildung der Normalkraft‐Momenten (N‐M)‐Beziehung und des Stabilitätsversagens ist eine Grundlage für die Entwicklung eines objektivierten Bemessungsansatzes und der zugehörigen Teilsicherheitsbeiwerte.

show abstract

“…A more accurate and a popular approximation method is the response surface method [1], where the LSF is approximated using a suitable function which most often is of the polynomial type [2]. Other methods which have gained considerations among researchers in the last decades include also artificial neural network [3], polynomial chaos [4], support vector machine [5] and the Kriging metamodel [6].…”

Section: Introductionmentioning

confidence: 99%

An adaptive ANN‐based inverse response surface method

Šomodíková

Lehký

2018

Beton und Stahlbetonbau

Self Cite

View full text Add to dashboard Cite

The paper deals with an adaptive artificial neural network‐based inverse response surface method utilized when performing reliability‐based design optimization of complex structural systems. Since calculating their reliability indicators (failure probabilities or reliability indices) is usually a time‐consuming task, the utilization of approximation methods with a view to reducing the computational effort to an acceptable level is an appropriate solution. A popular approximation method is the response surface method, where the limit state function is approximated using a suitable surrogate model. In this case, an artificial neural network is utilized. Construction of a response surface requires all variables of stochastic model to be known in advance. However, during the structural design, which is an inverse task, there are design parameters which are subject of reliability‐based design optimization procedure and thus not known at the start of the process. For such cases, an adaptive inverse response surface procedure is proposed. The procedure is based on a coupling of the adaptive response surface method and the artificial neural network‐based inverse reliability method. The validity and accuracy of the method is tested using example with explicit nonlinear limit state function. Obtained results as well as important aspects of the method are discussed.

show abstract

Reliability calculation of time-consuming problems using a small-sample artificial neural network-based response surface method

Cited by 34 publications

References 21 publications

A comparison of sensitivity analyses for selected prestressed concrete structures

A comparison of sensitivity analyses for selected prestressed concrete structures

Zuverlässigkeit schlanker Betonstützen

An adaptive ANN‐based inverse response surface method

Contact Info

Product

Resources

About