Application of LSTM based on the BAT-MCS for binary-state network approximated time-dependent reliability problems
•Propose the time-series reliability of the binary-state network problem.•Propose the first deep-learning algorithm to solve the problem.•The first algorithm combined the exact-algorithm and approximated-reliability algorithm for the problem.•The propose algorithm is based on LSTM, BAT, and MCS.•Val...
Saved in:
| Published in: | Reliability engineering & system safety Vol. 235; p. 108954 |
|---|---|
| Main Authors: | , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier Ltd
01.07.2023
|
| Subjects: | |
| ISSN: | 0951-8320, 1879-0836 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | •Propose the time-series reliability of the binary-state network problem.•Propose the first deep-learning algorithm to solve the problem.•The first algorithm combined the exact-algorithm and approximated-reliability algorithm for the problem.•The propose algorithm is based on LSTM, BAT, and MCS.•Validated on 3 larger-scale binary networks.
Reliability is an important tool for evaluating the performance of modern networks. Currently, it is NP-hard and #P-hard to calculate the exact reliability of a binary-state network when the reliability of each component is assumed to be fixed. However, this assumption is unrealistic because the reliability of each component always varies with time. To meet this practical requirement, we propose a new algorithm called the binary-addition-tree algorithm and Monte Carlo simulation based Long Short-Term Memory (LSTM-BAT-MCS), based on long short-term memory (LSTM), the Monte Carlo simulation (MCS), and the binary-addition-tree algorithm (BAT). The superiority of the proposed LSTM-BAT-MCS was demonstrated by experimental results of three benchmark networks with at most 10−4 mean square error. |
|---|---|
| ISSN: | 0951-8320 1879-0836 |
| DOI: | 10.1016/j.ress.2022.108954 |