Monte–carlo based method for predicting extreme value statistics of uncertain structures / Nilanjan Saha in Journal of engineering mechanics, Vol. 136 N° 12 (Décembre 2010)  

Public ISBD [article]  inJournal of engineering mechanics > Vol. 136 N° 12 (Décembre 2010) . - pp.1491-1501 Titre : Monte–carlo based method for predicting extreme value statistics of uncertain structures Type de document : texte imprimé Auteurs : Nilanjan Saha, Auteur ; Arvid Naess, Auteur Année de publication : 2011 Article en page(s) : pp.1491-1501 Note générale : Mécanique appliquée Langues : Anglais (eng) Mots-clés : Monte Carlo method Uncertainty principles Stochastic processes Finite element Turbines Predictions Excitation. Résumé : In the present paper, a simple method is proposed for predicting the extreme response of uncertain structures subjected to stochastic excitation. Many of the currently used approaches to extreme response predictions are based on the asymptotic generalized extreme value distribution, whose parameters are estimated from the observed data. However, in most practical situations, it is not easy to ascertain whether the given response time series contain data above a high level that are truly asymptotic, and hence the obtained parameter values by the adopted estimation methods, which points to the appropriate extreme value distribution, may become inconsequential. In this paper, the extreme value statistics are predicted taking advantage of the regularity of the tail region of the mean upcrossing rate function. This method is instrumental in handling combined uncertainties associated with nonergodic processes (system uncertainties) as well as ergodic ones (stochastic loading). For the specific applications considered, it can be assumed that the considered time series has an extreme value distribution that has the Gumbel distribution as its asymptotic limit. The present method is numerically illustrated through applications to a beam with spatially varying random properties and wind turbines subjected to stochastic loading. DEWEY : 620.1 ISSN : 0733-9399 En ligne : http://ascelibrary.org/emo/resource/1/jenmdt/v136/i12/p1491_s1?isAuthorized=no [article] Monte–carlo based method for predicting extreme value statistics of uncertain structures [texte imprimé] / Nilanjan Saha, Auteur ; Arvid Naess, Auteur . - 2011 . - pp.1491-1501. Mécanique appliquée Langues : Anglais (eng) in Journal of engineering mechanics > Vol. 136 N° 12 (Décembre 2010) . - pp.1491-1501 Mots-clés : Monte Carlo method Uncertainty principles Stochastic processes Finite element Turbines Predictions Excitation. Résumé : In the present paper, a simple method is proposed for predicting the extreme response of uncertain structures subjected to stochastic excitation. Many of the currently used approaches to extreme response predictions are based on the asymptotic generalized extreme value distribution, whose parameters are estimated from the observed data. However, in most practical situations, it is not easy to ascertain whether the given response time series contain data above a high level that are truly asymptotic, and hence the obtained parameter values by the adopted estimation methods, which points to the appropriate extreme value distribution, may become inconsequential. In this paper, the extreme value statistics are predicted taking advantage of the regularity of the tail region of the mean upcrossing rate function. This method is instrumental in handling combined uncertainties associated with nonergodic processes (system uncertainties) as well as ergodic ones (stochastic loading). For the specific applications considered, it can be assumed that the considered time series has an extreme value distribution that has the Gumbel distribution as its asymptotic limit. The present method is numerically illustrated through applications to a beam with spatially varying random properties and wind turbines subjected to stochastic loading. DEWEY : 620.1 ISSN : 0733-9399 En ligne : http://ascelibrary.org/emo/resource/1/jenmdt/v136/i12/p1491_s1?isAuthorized=no

Exemplaires

Code-barres	Cote	Support	Localisation	Section	Disponibilité
aucun exemplaire

Two-stage extended kalman filters with derivative-free local linearizations / Nilanjan Saha in Journal of engineering mechanics, Vol. 137 N° 8 (Août 2011)  

Public ISBD [article]  inJournal of engineering mechanics > Vol. 137 N° 8 (Août 2011) . - pp.537546 Titre : Two-stage extended kalman filters with derivative-free local linearizations Type de document : texte imprimé Auteurs : Nilanjan Saha, Auteur ; D. Roy, Auteur Année de publication : 2011 Article en page(s) : pp.537546 Note générale : Mécanique appliquée Langues : Anglais (eng) Mots-clés : Two-stage filter Extended Kalman filters Model uncertainty Derivative-free local linearizations Parameter estimations Résumé : This paper proposes a derivative-free two-stage extended Kalman filter (2-EKF) especially suited for state and parameter identification of mechanical oscillators under Gaussian white noise. Two sources of modeling uncertainties are considered: (1) errors in linearization, and (2) an inadequate system model. The state vector is presently composed of the original dynamical/parameter states plus the so-called bias states accounting for the unmodeled dynamics. An extended Kalman estimation concept is applied within a framework predicated on explicit and derivative-free local linearizations (DLL) of nonlinear drift terms in the governing stochastic differential equations (SDEs). The original and bias states are estimated by two separate filters; the bias filter improves the estimates of the original states. Measurements are artificially generated by corrupting the numerical solutions of the SDEs with noise through an implicit form of a higher-order linearization. Numerical illustrations are provided for a few single- and multidegree-of-freedom nonlinear oscillators, demonstrating the remarkable promise that 2-EKF holds over its more conventional EKF-based counterparts. DEWEY : 620.1 ISSN : 0733-9399 En ligne : http://ascelibrary.org/emo/resource/1/jenmdt/v137/i8/p537_s1?isAuthorized=no [article] Two-stage extended kalman filters with derivative-free local linearizations [texte imprimé] / Nilanjan Saha, Auteur ; D. Roy, Auteur . - 2011 . - pp.537546. Mécanique appliquée Langues : Anglais (eng) in Journal of engineering mechanics > Vol. 137 N° 8 (Août 2011) . - pp.537546 Mots-clés : Two-stage filter Extended Kalman filters Model uncertainty Derivative-free local linearizations Parameter estimations Résumé : This paper proposes a derivative-free two-stage extended Kalman filter (2-EKF) especially suited for state and parameter identification of mechanical oscillators under Gaussian white noise. Two sources of modeling uncertainties are considered: (1) errors in linearization, and (2) an inadequate system model. The state vector is presently composed of the original dynamical/parameter states plus the so-called bias states accounting for the unmodeled dynamics. An extended Kalman estimation concept is applied within a framework predicated on explicit and derivative-free local linearizations (DLL) of nonlinear drift terms in the governing stochastic differential equations (SDEs). The original and bias states are estimated by two separate filters; the bias filter improves the estimates of the original states. Measurements are artificially generated by corrupting the numerical solutions of the SDEs with noise through an implicit form of a higher-order linearization. Numerical illustrations are provided for a few single- and multidegree-of-freedom nonlinear oscillators, demonstrating the remarkable promise that 2-EKF holds over its more conventional EKF-based counterparts. DEWEY : 620.1 ISSN : 0733-9399 En ligne : http://ascelibrary.org/emo/resource/1/jenmdt/v137/i8/p537_s1?isAuthorized=no

Exemplaires

Code-barres	Cote	Support	Localisation	Section	Disponibilité
aucun exemplaire