What question did this study set out to answer?

The research aims to improve fault diagnosis in rotating machinery using a few-shot learning approach that addresses weak features and generalization issues.

May 29, 2026Open Access

Research on few-shot fault diagnosis and feature extraction mechanism based on ADPCW-ELCNN

Puntos clave

The research aims to improve fault diagnosis in rotating machinery using a few-shot learning approach that addresses weak features and generalization issues.
Proposed a novel Adaptive Dual-Parameter Collaborative Wavelet Convolutional Neural Network for fault diagnosis.
Tested the framework on bearing and gearbox datasets with a focus on weak feature extraction under noise conditions.
Investigated model performance emphasizing parameter collaboration and feature adaptability.
Achieved 99.98% average accuracy with 9.19 K parameters and 0.02 × 103 M FLOPs.
Maintained F1-score over 94% in low-sample (20 samples) and noisy (-10 dB SNR) conditions.

Resumen

Aiming at weak features masked by noise, disconnected parameter-feature adaptation, and unbalanced generalization-efficiency in rotating machinery few-shot fault diagnosis, an Adaptive Dual-Parameter Collaborative Wavelet Convolutional Neural Network is proposed. Different from existing frameworks relying on data augmentation, transfer learning, or blind structural optimization, it adopts a physics-informed, parameter-collaborative, feature-faithful paradigm. Core innovations include quantitative correlation between wavelet scale and convolution kernel length, global–local adaptive strategy, and cascaded nonlinear enhancement. Experimental results on bearing and gearbox datasets show the model has 9.19 K parameters and 0.02 × 103 M FLOPs, with 99.98% average accuracy. It maintains F1-score over 94% under 20-sample and − 10 dB SNR conditions, providing an efficient solution for limited-sample fault diagnosis.

Me gusta

Guardar

Ver artículo completo