Research on Fault Detection Method for Aircraft Key Systems Based on Graph Convolutional Networks
Year: 2025
Type: conference-paper
Abstract: This paper presents a fault detection approach for aircraft key systems, which is based on Graph Convolutional Network(GCN). In the process of collecting various key parameter data, which is done at high frequencies during the operations of aircraft, a certain structure in the form of a graph is created, and node features are defined. By the method of graph convolutional network, important messages are passed and features are updated, resulting in the model being able to detect and evaluate potential faults in aircraft engine systems and flight attitude systems. Experimental results show its effectiveness in identifying potential aircraft faults, thereby providing robust support for ensuring flight safety.
Authors: Peili Liu, Wenquan Wang, Dong Ning, Yang Liu
Institution: Liaocheng University
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4413845112