西安建筑科技大学教师个人主页 Lichen Shi--Home-- Small-sample gear fault diagnosis method based on GASF and MSCAM-DenseNet

Small-sample gear fault diagnosis method based on GASF and MSCAM-DenseNet

Release time:2025-02-15

Hits:

Key Words:: gear; small-sample fault diagnosis; Gramian Angular Summation Field; Two-dimensional discrete wavelet transform; multi-scale channel attention mechanism

Abstract:: To address the problem of insufficient samples obtained under small sample conditions and the decrease in diagnostic accuracy caused by ineffective feature extraction, a small sample gear fault diagnosis method combining GASF (Gramian Angular Summation Field) and MSCAM-DenseNet is proposed. Firstly, the Gramian Angular Summation Field (GASF) is used to transform the multi-source vibration signals into two-dimensional features, and the multi-source features are reconstructed using two-dimensional discrete wavelet transform (2D-DWT). Secondly, since the conventional DenseNet lacks the ability to recognize multi-scale features, a multi-scale channel attention mechanism (MSCAM) is introduced into DenseNet, proposing an improved network model called MSCAM-DenseNet. Finally, the reconstructed GASF is used as the input to MSCAM-DenseNet, and after feature recognition is completed, a network classifier is used to classify the fault features. The proposed model is validated using the planetary gear dataset obtained from the laboratory and the gearbox dataset from Southeast University, and compared with other diagnostic models. Experimental results demonstrate that the proposed method achieves high fault recognition accuracy, strong generalization ability, and noise resistance under small-sample and varying operating conditions.

Links to published journals:: https://link.cnki.net/urlid/11.5946.TP.20231115.1155.004