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

2026-04-02

AI Summary

This study addresses the challenges of limited computational resources and insufficient interpretability in general aviation fault diagnosis on edge devices by proposing LiteInception, a two-stage cascaded framework. The first stage prioritizes high-recall fault detection, while the second performs fine-grained anomaly classification. The method innovatively integrates mutual information, gradient-based analysis, and squeeze-and-excitation (SE) attention for sensor channel compression, combined with knowledge distillation to optimize the trade-off between precision and recall. A dual-layer interpretability framework—spanning “sensor × time window”—is introduced, unifying four attribution techniques. Evaluated on the NGAFID dataset, the model achieves a 70% reduction in parameters, over 8× faster CPU inference, and less than 3% F1-score degradation, delivering 81.92% fault detection accuracy (with 83.24% recall) and 77.00% fault identification accuracy.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Computer Vision: Interpretability, Explainability, and Transparency

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Machine-in-the-loop, human agency and autonomy

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.

Problem

• fault diagnosis
• edge computing
• model interpretability
• lightweight deep learning
• general aviation

Innovation

• LiteInception
• edge deployment
• model compression
• knowledge distillation
• interpretability