LiteInception: A Lightweight and Interpretable Deep Learning Framework for General Aviation Fault Diagnosis

TL;DR

LiteInception is a lightweight, interpretable deep learning framework tailored for fault diagnosis in general aviation, optimized for resource-constrained edge devices with high accuracy and explainability.

Contribution

The paper introduces a novel cascaded architecture, multi-method model compression, knowledge distillation, and a dual-layer interpretability framework for efficient aviation fault diagnosis.

Findings

Fault detection accuracy of 81.92% with 83.24% recall.

Model reduces parameters by 70% and accelerates inference by over 8x.

Framework achieves a good balance of efficiency, accuracy, and interpretability.

Abstract

General aviation fault diagnosis and efficient maintenance are critical to flight safety; however, deploying deep learning models on resource-constrained edge devices poses dual challenges in computational capacity and interpretability. This paper proposes LiteInception--a lightweight interpretable fault diagnosis framework designed for edge deployment. The framework adopts a two-stage cascaded architecture aligned with standard maintenance workflows: Stage 1 performs high-recall fault detection, and Stage 2 conducts fine-grained fault classification on anomalous samples, thereby decoupling optimization objectives and enabling on-demand allocation of computational resources. For model compression, a multi-method fusion strategy based on mutual information, gradient analysis, and SE attention weights is proposed to reduce the input sensor channels from 23 to 15, and a 1+1 branch LiteInception architecture is introduced that compresses InceptionTime parameters by 70%, accelerates CPU inference by over 8x, with less than 3% F1 loss. Furthermore, knowledge distillation is introduced as a precision-recall regulation mechanism, enabling the same lightweight model to adapt to different scenarios--such as safety-critical and auxiliary diagnosis--by switching training strategies. Finally, a dual-layer interpretability framework integrating four attribution methods is constructed, providing traceable evidence chains of "which sensor x which time period." Experiments on the NGAFID dataset demonstrate a fault detection accuracy of 81.92% with 83.24% recall, and a fault identification accuracy of 77.00%, validating the framework's favorable balance among efficiency, accuracy, and interpretability.