Intelligent fault diagnosis method of planetary gearboxes based on convolution neural network and discrete wavelet transform

• Published in: Computers in industry 2019-04, Vol.106, p.48-59
• Main Authors: Chen, Renxiang, Huang, Xin, Yang, Lixia, Xu, Xiangyang, Zhang, Xia, Zhang, Yong
• Format: Article
• Language: English
• Subjects: Convolution neural network; Discrete wavelet transformation; Fault diagnosis; Feature extraction; Planetary gearboxes
• ISSN: 0166-3615; 1872-6194
• DOI: 10.1016/j.compind.2018.11.003

•The paper focus on feature extraction and classification using deep learning method from time-frequency domain.•Based on exploring the advantage of different signal processing methods, we combined the deep learning method CNN with time-frequency domain signal processing method DWT to construct fault diagnosis model.•A series of comparative experiment with known conventional fault diagnosis solutions verified the validity of our proposed method in noiseless and noisy environment.•The proposed method achieved highly accuracy in noiseless and noisy environment, indicating the method had strongly denoising ability, and feature extraction and fault recognition were incorporated into a general-purpose learning procedure. Considering the planetary gearbox vibration signals show highly non-stationary and non-linear behavior because of wind turbines (WTs) often working under time-varying running conditions, we propose an effective and reliable method based on convolution neural network (CNN) and discrete wavelet transformation (DWT) to identify the fault conditions of planetary gearboxes. Firstly, the discrete wavelet transformation is used with the goal of presenting more salient and comprehensive time-frequency distributed representation. Secondly, the deep hierarchical structure of CNN constructed by the alternating convolution layers and subsample layers is trained using a forward transmitting rule of greedy training layer by layer and translates the low-level features of input to the high-level features in order to identify the internal characteristic. Finally, a top classifier Softmax is added at uppermost layer of CNN and the backpropagation process is conducted to fine-tune the parameters of CNN, establishing the mapping relation among the feature space and the fault space. Thus, feature extraction process and fault recognition are incorporated into a general-purpose learning procedure. The experimental results of the fault diagnosis for the planetary gearbox demonstrated the effectiveness and feasibility of the proposed method.