LiteInception: A Lightweight and Interpretable Deep Learning Framework for General Aviation Fault Diagnosis
This work addresses the problem of efficient and interpretable fault diagnosis for general aviation maintenance, though it appears incremental as it builds on existing methods like InceptionTime with optimizations for edge deployment.
The paper tackles the challenge of deploying deep learning models for general aviation fault diagnosis on resource-constrained edge devices by proposing LiteInception, a lightweight and interpretable framework that reduces parameters by 70%, accelerates CPU inference by over 8x with less than 3% F1 loss, and achieves fault detection accuracy of 81.92% with 83.24% recall.
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.