Knowledge Graph Construction Method for Commercial Aircraft Fault Diagnosis Based on Logic Diagram Model
Commercial aircraft fault diagnosis is an important means to ensure the reliability and safety of commercial aircraft. Traditional knowledge-driven and data-driven fault diagnosis methods lack interpretability in engineering mechanisms, making them difficult to promote and apply. To address the issu...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-09-01
|
| Series: | Aerospace |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2226-4310/11/9/773 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Commercial aircraft fault diagnosis is an important means to ensure the reliability and safety of commercial aircraft. Traditional knowledge-driven and data-driven fault diagnosis methods lack interpretability in engineering mechanisms, making them difficult to promote and apply. To address the issue of lack of interpretability, this paper conducts a fault knowledge graph for commercial aircraft fault diagnosis, using the fault logic in the logic diagram to increase the interpretability of diagnostic work. Firstly, to avoid the inefficiency of logic diagram applications, an executable logic diagram model is established, which can perform mathematical analysis and achieve fault diagnosis and localization using operational data as input. Then, the logic diagram is sorted out to obtain the hidden fault knowledge in the logic diagram, which is used to construct a fault knowledge graph to help achieve cause localization and rapid troubleshooting. The methods proposed in this paper are all validated through case studies of abnormal low-pressure faults in domestic commercial aircraft hydraulic systems. The results show that the logic diagram model can perform model simulation and fault diagnosis based on operational data, and the fault knowledge graph can quickly locate abnormal monitoring parameters and guide troubleshooting work based on existing information. |
|---|---|
| ISSN: | 2226-4310 |