Details
| Originalsprache | Englisch |
|---|---|
| Fachzeitschrift | UNCECOMP Proceedings |
| Jahrgang | 2021 |
| Publikationsstatus | Veröffentlicht - 2021 |
| Veranstaltung | 4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering - online, Athens, Griechenland Dauer: 28 Juni 2021 → 30 Juni 2021 Konferenznummer: 4 |
Abstract
In order to describe spatially uncertain parameters by random fields, the underlying autocorrelation structure in engineering structures is usually not known.. The idea of imprecise random fields is to acknowledge this lack of knowledge by adding epistemic uncertainties. Within this contribution the influence of the correlation length is studied. In particular, it is shown that there exist bounds that limit the case of having no idea at all. This “absolutely no idea p-box” is defined by white noise and the random variable corresponding to the mean value and standard deviation of the imprecise random field. By this, the limits of having “absolutely no idea” can be described without the need of Karhunen-Loève expansion and random field propagation. Then, at least for linear problems, every response in between can be estimated by linear interpolation without any need for sampling.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Theoretische Informatik und Mathematik
- Informatik (insg.)
- Angewandte Informatik
- Mathematik (insg.)
- Modellierung und Simulation
- Mathematik (insg.)
- Statistik und Wahrscheinlichkeit
- Mathematik (insg.)
- Steuerung und Optimierung
- Mathematik (insg.)
- Diskrete Mathematik und Kombinatorik
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
in: UNCECOMP Proceedings, Jahrgang 2021, 2021.
Publikation: Beitrag in Fachzeitschrift › Konferenzaufsatz in Fachzeitschrift › Forschung › Peer-Review
}
TY - JOUR
T1 - Limit representations of imprecise random fields
AU - Dannert, Mona M.
AU - Häufler, Johannes L.
AU - Nackenhorst, Udo
N1 - Conference code: 4
PY - 2021
Y1 - 2021
N2 - In order to describe spatially uncertain parameters by random fields, the underlying autocorrelation structure in engineering structures is usually not known.. The idea of imprecise random fields is to acknowledge this lack of knowledge by adding epistemic uncertainties. Within this contribution the influence of the correlation length is studied. In particular, it is shown that there exist bounds that limit the case of having no idea at all. This “absolutely no idea p-box” is defined by white noise and the random variable corresponding to the mean value and standard deviation of the imprecise random field. By this, the limits of having “absolutely no idea” can be described without the need of Karhunen-Loève expansion and random field propagation. Then, at least for linear problems, every response in between can be estimated by linear interpolation without any need for sampling.
AB - In order to describe spatially uncertain parameters by random fields, the underlying autocorrelation structure in engineering structures is usually not known.. The idea of imprecise random fields is to acknowledge this lack of knowledge by adding epistemic uncertainties. Within this contribution the influence of the correlation length is studied. In particular, it is shown that there exist bounds that limit the case of having no idea at all. This “absolutely no idea p-box” is defined by white noise and the random variable corresponding to the mean value and standard deviation of the imprecise random field. By this, the limits of having “absolutely no idea” can be described without the need of Karhunen-Loève expansion and random field propagation. Then, at least for linear problems, every response in between can be estimated by linear interpolation without any need for sampling.
KW - Imprecise random fields
KW - Interval valued correlation length
KW - Karhunen-Loève expansion
KW - Stochastic finite element method
UR - http://www.scopus.com/inward/record.url?scp=85121105943&partnerID=8YFLogxK
U2 - 10.7712/120221.8024.19110
DO - 10.7712/120221.8024.19110
M3 - Conference article
AN - SCOPUS:85121105943
VL - 2021
JO - UNCECOMP Proceedings
JF - UNCECOMP Proceedings
T2 - 4th International Conference on Uncertainty Quantification in Computational Sciences and Engineering, UNCECOMP 2021
Y2 - 28 June 2021 through 30 June 2021
ER -