A Novel Deep Model with Meta-learning for Rolling Bearing Few-shot Fault Diagnosis
Xiaoxia Liang, Ming Zhang, Guojin Feng, Yuchun Yu, Dong Zhen, Fengshou Gu
Abstract
Machine learning, especially deep learning, has been highly successful in data-intensive applications, however, the performance of these models will drop significantly when the amount of the training data amount does not meet the requirement. This leads to the so-called Few-Shot Learning (FSL) problem, which requires the model rapidly generalize to new tasks that containing only a few labeled samples. In this paper, we proposed a new deep model, called deep convolutional meta-learning networks (DCMLN), to address the low performance of generalization under limited data for bearing fault diagnosis. The essential of our approach is to learn a base model from the multiple learning tasks using a support dataset and finetune the learnt parameters using few-shot tasks before it can adapt to the new learning task based on limited training data. The proposed method was compared to several few-shot learning methods, including methods with and without pre-training the embedding mapping, and methods with finetuning the classifier or the whole model by utilizing the few-shot data from the target domain. The comparisons are carried out on one-shot and ten-shot tasks using the CWRU bearing dataset and a cylindrical roller bearing dataset. The experimental result illustrates that our method has good performance on the bearing fault diagnosis across various few-shot conditions. In addition, we found that the pre-training process does not always improve the prediction accuracy.