Multitask Learning Based on Lightweight 1DCNN for Fault Diagnosis of Wheelset Bearings

Abstract

In recent years, deep learning has been proved to be a promising bearing fault diagnosis technology. However, most of the existing methods are based on single-task learning. Fault diagnosis task (FDT) is treated as an independent task, and rich correlation information contained in different tasks is ignored. Therefore, this article explores the possibility of using speed identification task (SIT) and load identification task (LIT) as two auxiliary tasks to improve the performance of the FDT and proposes a multitask one-dimensional convolutional neural network (MT-1DCNN). Specifically, the MT-1DCNN utilizes trunk network to learn shared features required for every task and then processes different tasks through multiple task-specific branches. In this way, the MT-1DCNN can utilize features learned by related tasks to improve the performance of the FDT. The experimental results with wheelset bearing data set show that the multitask learning can make full use of the feature information captured by the SIT and the LIT to improve the fault diagnosis performance of the network, and the MT-1DCNN has a better performance than five excellent networks in accuracy.