Details
| Originalsprache | Englisch |
|---|---|
| Aufsatznummer | 107935 |
| Fachzeitschrift | Reliability engineering & system safety |
| Jahrgang | 216 |
| Frühes Online-Datum | 24 Juli 2021 |
| Publikationsstatus | Veröffentlicht - Dez. 2021 |
Abstract
Societal growth thrives on the performance of critical infrastructure systems such as water supply systems, transportation networks or electrical distribution systems. This makes the reliability analysis of these systems a core focus for researchers today. The survival signature is a novel tool for analysing complex networks efficiently and outperforms traditional techniques in several key factors. Its most unique feature being a full separation of the system structure from probabilistic information. This in turn allows for the consideration of diverse component failure descriptions such as dependencies, common causes of failure and imprecise probabilities. However, the numerical effort to compute the survival signature increases with network size and prevents analysis of complex systems. This work presents a new method to approximate the survival signature, where system configurations of low interest are first excluded using percolation theory, while the remaining parts of the signature are approximated by Monte Carlo simulation. The approach is able to accurately approximate the survival signature with very small error at a massive reduction in computational demands. The accuracy and performance are highlighted using several simple test systems as well as two real world problems.
ASJC Scopus Sachgebiete
- Ingenieurwesen (insg.)
- Sicherheit, Risiko, Zuverlässigkeit und Qualität
- Ingenieurwesen (insg.)
- Wirtschaftsingenieurwesen und Fertigungstechnik
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
in: Reliability engineering & system safety, Jahrgang 216, 107935, 12.2021.
Publikation: Beitrag in Fachzeitschrift › Artikel › Forschung › Peer-Review
}
TY - JOUR
T1 - Numerically efficient computation of the survival signature for the reliability analysis of large networks
AU - Behrensdorf, Jasper
AU - Regenhardt, Tobias-Emanuel
AU - Broggi, Matteo
AU - Beer, Michael
N1 - Funding Information: The research work herein was supported by the German Research Foundation under Grant No. BE 2570/3–1 and BR 5446/1–1 . This support is gratefully acknowledged.
PY - 2021/12
Y1 - 2021/12
N2 - Societal growth thrives on the performance of critical infrastructure systems such as water supply systems, transportation networks or electrical distribution systems. This makes the reliability analysis of these systems a core focus for researchers today. The survival signature is a novel tool for analysing complex networks efficiently and outperforms traditional techniques in several key factors. Its most unique feature being a full separation of the system structure from probabilistic information. This in turn allows for the consideration of diverse component failure descriptions such as dependencies, common causes of failure and imprecise probabilities. However, the numerical effort to compute the survival signature increases with network size and prevents analysis of complex systems. This work presents a new method to approximate the survival signature, where system configurations of low interest are first excluded using percolation theory, while the remaining parts of the signature are approximated by Monte Carlo simulation. The approach is able to accurately approximate the survival signature with very small error at a massive reduction in computational demands. The accuracy and performance are highlighted using several simple test systems as well as two real world problems.
AB - Societal growth thrives on the performance of critical infrastructure systems such as water supply systems, transportation networks or electrical distribution systems. This makes the reliability analysis of these systems a core focus for researchers today. The survival signature is a novel tool for analysing complex networks efficiently and outperforms traditional techniques in several key factors. Its most unique feature being a full separation of the system structure from probabilistic information. This in turn allows for the consideration of diverse component failure descriptions such as dependencies, common causes of failure and imprecise probabilities. However, the numerical effort to compute the survival signature increases with network size and prevents analysis of complex systems. This work presents a new method to approximate the survival signature, where system configurations of low interest are first excluded using percolation theory, while the remaining parts of the signature are approximated by Monte Carlo simulation. The approach is able to accurately approximate the survival signature with very small error at a massive reduction in computational demands. The accuracy and performance are highlighted using several simple test systems as well as two real world problems.
KW - Monte Carlo simulation
KW - Percolation
KW - Reliability analysis
KW - Survival signature
UR - http://www.scopus.com/inward/record.url?scp=85111499379&partnerID=8YFLogxK
U2 - 10.1016/j.ress.2021.107935
DO - 10.1016/j.ress.2021.107935
M3 - Article
AN - SCOPUS:85111499379
VL - 216
JO - Reliability engineering & system safety
JF - Reliability engineering & system safety
SN - 0951-8320
M1 - 107935
ER -