Details
| Originalsprache | Englisch |
|---|---|
| Seitenumfang | 8 |
| Publikationsstatus | Veröffentlicht - 26 Mai 2019 |
| Veranstaltung | 13th International Conference on Applications of Statistics and Probability in Civil Engineering - Seoul, South Korea, Seoul, Südkorea Dauer: 26 Mai 2019 → 30 Mai 2019 Konferenznummer: 13 |
Konferenz
| Konferenz | 13th International Conference on Applications of Statistics and Probability in Civil Engineering |
|---|---|
| Kurztitel | ICASP13 |
| Land/Gebiet | Südkorea |
| Ort | Seoul |
| Zeitraum | 26 Mai 2019 → 30 Mai 2019 |
Abstract
Structural reliability analysis is typically performed based on the identification of distribution types of random inputs. However, this is often not feasible in engineering practice due to limited available probabilistic information (e.g., limited observed samples or physics-based inference). In this paper, a linear programming-based approach is developed to perform structural reliability analysis subjected to incompletely informed random variables. The approach converts a reliability analysis into a standard linear programming problem, which can make full use of the probabilistic information of the variables. The proposed method can also be used to construct the best-possible distribution function bounds for a random variable with limited statistical information. Illustrative examples are presented to demonstrate the applicability and efficiency of the proposed method. It is shown that the proposed approach can provide a tighter estimate of structural reliability bounds compared with existing interval Monte Carlo methods which propagate probability boxes.
ASJC Scopus Sachgebiete
- Ingenieurwesen (insg.)
- Tief- und Ingenieurbau
- Mathematik (insg.)
- Statistik und Wahrscheinlichkeit
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
2019. Beitrag in 13th International Conference on Applications of Statistics and Probability in Civil Engineering, Seoul, Südkorea.
Publikation: Konferenzbeitrag › Paper › Forschung › Peer-Review
}
TY - CONF
T1 - Tightening the bound estimate of structural reliability under imprecise probability information
AU - Wang, Cao
AU - Zhang, Hao
AU - Beer, Michael
N1 - Conference code: 13
PY - 2019/5/26
Y1 - 2019/5/26
N2 - Structural reliability analysis is typically performed based on the identification of distribution types of random inputs. However, this is often not feasible in engineering practice due to limited available probabilistic information (e.g., limited observed samples or physics-based inference). In this paper, a linear programming-based approach is developed to perform structural reliability analysis subjected to incompletely informed random variables. The approach converts a reliability analysis into a standard linear programming problem, which can make full use of the probabilistic information of the variables. The proposed method can also be used to construct the best-possible distribution function bounds for a random variable with limited statistical information. Illustrative examples are presented to demonstrate the applicability and efficiency of the proposed method. It is shown that the proposed approach can provide a tighter estimate of structural reliability bounds compared with existing interval Monte Carlo methods which propagate probability boxes.
AB - Structural reliability analysis is typically performed based on the identification of distribution types of random inputs. However, this is often not feasible in engineering practice due to limited available probabilistic information (e.g., limited observed samples or physics-based inference). In this paper, a linear programming-based approach is developed to perform structural reliability analysis subjected to incompletely informed random variables. The approach converts a reliability analysis into a standard linear programming problem, which can make full use of the probabilistic information of the variables. The proposed method can also be used to construct the best-possible distribution function bounds for a random variable with limited statistical information. Illustrative examples are presented to demonstrate the applicability and efficiency of the proposed method. It is shown that the proposed approach can provide a tighter estimate of structural reliability bounds compared with existing interval Monte Carlo methods which propagate probability boxes.
UR - http://www.scopus.com/inward/record.url?scp=85083952223&partnerID=8YFLogxK
U2 - 10.22725/ICASP13.248
DO - 10.22725/ICASP13.248
M3 - Paper
T2 - 13th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP 2019
Y2 - 26 May 2019 through 30 May 2019
ER -