1-D Convolutional Neural Network for Integrated Modular Avionics Fault Modeling and Diagnosis

Inhalt:
With the development of integrated avionics, integrated modular avionics (IMA) system has become an important part of modern aircraft airborne systems. Once IMA fails, it will not only affect the operation of avionics functions, but also threaten flight safety. Therefore, it is necessary to carry out fault modeling and diagnosis, which can identify its failure cause accurately and improve its maintenance efficiency. Based on the IMA failure mechanism analysis and fault data, the 1-D Convolutional Neural Network is introduced in this paper to model the IMA system’s failure process and determine its failure cause, the effectiveness of the proposed method for fault modeling and diagnosis is verified by the experimental simulation in the end.