Manifold-Contrastive Broad Learning System for Wheelset Bearing Fault Diagnosis

Newly deployed trains have massive normal data and scarce faulty data for training, which limits the diagnosis accuracy with class imbalance problem of small samples. Considering that there are a lot unutilized information hidden in the abundant unlabeled monitoring data, this paper proposes a novel method named manifold-contrastive broad learning system, which utilizes the online updating approach for dealing with the class imbalance problem of small samples. This method constructs a novel one-class broad-learning classifier based on an inherency-guided comparison mechanism, which can classify and annotate unlabeled data online. This classifier employs contrastive manifold matrices to maintain the inherent structures, which is not affected to the overfitting caused by imbalanced samples. Secondly, inspired by the active learning, this classifier proposes the minimum-error strategy to annotate the samples by classifying the modes, which solves the problem of insufficient training data. Thirdly, this method applies an incremental learning strategy that continuously absorbs the newly annotated data to update the model online, which improves the model accuracy under the data imbalanced condition. Finally, the feasibility and effectiveness of the proposed method are verified by wheelset bearing data collected from a test rig of a Chinese rolling stock company.

I. Introduction

High-speed trains have become more and more important in traffic and transportation due to their high speed, stability, and convenience [1], [2]. However, a bullet train is a complex system consisting of numerous close-coupled components [3], [4], whose safety requirements are extremely high. Therefore, it is essential to diagnose faults for newly deployed trains in time based on massive monitoring data [5]. However, since the collected monitoring data of new trains are limited [6], [7], especially the faulty data are rare, it is difficult to develop an accurate enough fault diagnosis model [8]. By now, a proven solution is to identify the monitoring data online and absorb the new information into the model for updating it synchronously.