Transfer Learning With Neural Networks for Bearing Fault Diagnosis in Changing Working Conditions

By Ran Zhang, Hongyan Tao, Lifeng Wu, Yong Guan. IEEE Access, 2017

333 citations recorded by the source corpus.

Traditional machine learning algorithms have made great achievements in data-driven fault diagnosis. However, they assume that all the data must be in the same working condition and have the same distribution and feature space. They are not applicable for real-world working conditions, which often change with time, so the data are hard to obtain. In order to utilize data in different working conditions to improve the performance, this paper presents a transfer learning approach for fault diagnosis with neural networks. First, it learns characteristics from massive source data and adjusts the parameters of neural networks accordingly. Second, the structure of neural networks alters for the change of data distribution. In the same time, some parameters are transferred from source task to target task. Finally, the new model is trained by a small amount of target data in another working condition. The Case Western Reserve University bearing data set is used to validate the performance of the proposed transfer learning approach. Experimental results show that the proposed transfer learning approach can improve the classification accuracy and reduce the training time comparing with the conventional neural network method when there are only a small amount of target data.

Subfields: Artificial Intelligence, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/ba21455ddcf5244ebd9b792afd144b8f17e7c416

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Shengli Zhang
2. Yaguo Lei
3. Shibin Wang
4. Xiaolei Yu
5. Tao Yan
6. Liang Guo
7. Qiyang Zhang
8. Shaobo Li

In Control and Systems Engineering

1. Yaguo Lei
2. Naipeng Li
3. Xuefeng Chen
4. Liang Guo
5. Ruqiang Yan
6. Tao Yan
7. Chuang Sun
8. Yihao Xue

Citations in the corpus

• Deep Transfer Learning for Bearing Fault Diagnosis: A Systematic Review Since 2016 (2023, methodology): “To resolve such a label inconsistency problem, the fine-tunebased methods were adopted by replacing the output layer of the pretrained model with a new one of the same dimension as target labels [108], [109].”
• Deep Convolutional Transfer Learning Network: A New Method for Intelligent Fault Diagnosis of Machines With Unlabeled Data (2019, background): “As a matter of fact, it is demonstrated in [8]–[10] that the success of intelligent fault diagnosis of machines relies on those two conditions.”
• Online Fault Diagnosis Method Based on Transfer Convolutional Neural Networks (2020, result): “Several previous studies [34]–[36] have drawn two conclusions: 1) initializing a deep neural network with transferred features learned from related data sets or tasks can boost the generalization performance and improve the efficiency of the network and 2) low-level features from shallow layers are ”
• Limited Data Rolling Bearing Fault Diagnosis With Few-Shot Learning (2019, background): “networks (RNNs) [25]–[27], transfer learning based neural networks [28]–[31].”
• Preprocessing-Free Gear Fault Diagnosis Using Small Datasets With Deep Convolutional Neural Network-Based Transfer Learning (2017, background): “Transfer learning, on the other hand, is capable of achieving prominent performance commensurate with large scale CNNs using only a small set of training date [31], [32].”
• A Digital-Twin-Assisted Fault Diagnosis Using Deep Transfer Learning (2019, background): “[36] presented a transfer learning approach for bearing fault diagnosis with neural networks, they used shallow structure rather than deep neural network to tackle the same issues as DTL.”
• Applications of Unsupervised Deep Transfer Learning to Intelligent Fault Diagnosis: A Survey and Comparative Study (2019, methodology): “2) Applications to IFD: Pre-trained deep neural networks using the source data were used in [26]–[37] via frozing their partial parameters, and then part of network parameters were transferred to the target network and other parameters were fine-tuned with a small amount of target data.”