A Semi-Supervised Fault Diagnosis Method Based on Improved Bidirectional Generative Adversarial Network

Long Cui,Xincheng Tian,Xiaorui Shi,Yigang Cui,Xiujing Wang

doi:10.3390/app11209401

Long Cui, Xincheng Tian + Show 3 more

Open Access

https://doi.org/10.3390/app11209401

Copy DOI

Abstract

With the assumption of sufficient labeled data, deep learning based machinery fault diagnosis methods show effectiveness. However, in real-industrial scenarios, it is costly to label the data, and unlabeled data is underutilized. Therefore, this paper proposes a semi-supervised fault diagnosis method called Bidirectional Wasserstein Generative Adversarial Network with Gradient Penalty (BiWGAN-GP). First, by unsupervised pre-training, the proposed method takes full advantage of a large amount of unlabeled data and can extract features from vibration signals effectively. Then, using only a few labeled data to conduct supervised fine-tuning, the model can perform an accurate fault diagnosis. Additionally, Wasserstein distance is used to improve the stability of the model’s training procedure. Validation is performed on the bearing and gearbox fault datasets with limited labeled data. The results show that the proposed method can achieve 99.42% and 91.97% of diagnosis accuracy on the bearing and gear dataset, respectively, when the size of the training set is only 10% of the testing set.

Highlights

Sinotruk Industry Park Zhangqiu, Sinotruk Jinan Power Co., Ltd., Jinan 250220, China; Abstract: With the assumption of sufficient labeled data, deep learning based machinery fault diagnosis methods show effectiveness
The results show that the loss of bidirectional generative adversarial network (BiGAN)
The fault diagnosis experiments demonstrate that the number of labeled samples is an essential factor in the performance of the semi-supervised learning model; the utilization of unlabeled samples has a significant impact on diagnostic accuracy

Summary

Introduction

Sinotruk Industry Park Zhangqiu, Sinotruk Jinan Power Co., Ltd., Jinan 250220, China; Abstract: With the assumption of sufficient labeled data, deep learning based machinery fault diagnosis methods show effectiveness. By unsupervised pre-training, the proposed method takes full advantage of a large amount of unlabeled data and can extract features from vibration signals effectively. The results show that the proposed method can achieve 99.42% and 91.97% of diagnosis accuracy on the bearing and gear dataset, respectively, when the size of the training set is only 10%. Deep learning based methods have attracted attention—because its ability to automatically extract features, which are less dependent on prior knowledge about signal processing techniques and diagnostic expertise [4]. With the assumption of sufficient data, supervised deep learning based fault diagnosis methods show effectiveness. Duong et al [6] used the wavelet transform to get the 2D image of the acoustic emission signals of the bearing, applied

Methods

Findings

Discussion

Conclusion