Robust Detection of Bearing Early Fault Based on Deep Transfer Learning

Abstract

In recent years, machine learning techniques have been proven to be a promising tool for early fault detection of rolling bearings. In many actual applications, however, bearing whole-life data are not easy to be historically accumulated, while insufficient data may result in training a detection model that is not good enough. If utilizing the available data under different working conditions to facilitate model training, the data distribution of different bearings are usually quite different, which does not meet the precondition of i n d e p e n d e n t a n d i d e n t i c a l d i s t r i b u t i o n ( i . i . d ) and tends to cause performance reduction. In addition, disturbed by the unstable noise under complex conditions, most of the current detection methods are inclined to raise false alarms, so that the reliability of detection results needs to be improved. To solve these problems, a robust detection method for bearings early fault is proposed based on deep transfer learning. The method includes offline stage and online stage. In the offline stage, by introducing a deep auto-encoder network with domain adaptation, the distribution inconsistency of normal state data among different bearings can be weakened, then the common feature representation of the normal state is obtained. With the extracted common features, a new state assessment method based on the robust deep auto-encoder network is proposed to evaluate the boundary between normal state and early fault state in the low-rank feature space. By training a support vector machine classifier, the detection model is established. In the online stage, along with the data batch arriving sequentially, the features of target bearing are extracted using the common representation learnt in the offline stage, and online detection is conducted by feeding them into the SVM model. Experimental results on IEEE PHM Challenge 2012 bearing dataset and XJTU-SY dataset show that the proposed approach outperforms several state-of-the-art detection methods in terms of detection accuracy and false alarm rate.

A New Feature Extraction Technique for Early Degeneration Detection of Rolling Bearings

“…However, kurtosis can just reflect the distribution density of shock signals, it ignores the components with large amplitude and scattered distribution. The correlation coefficient can express the reproduction degree of the component to the original signal, it is easy to be disturbed by noise [31]. Obviously, a single index can not fully interpret the vibration characteristics of the signal.…”

“…Step5: Calculate the EHNR value of the reconstructed signal. According to the 4 criterion of Gaussian distribution [31], Set the alarm threshold to the relatively stable EHNR mean value  in the normal period plus 4 times the standard deviation  , and the data beyond this interval is the abnormal point. Judge whether the bearing enters the degradation period according to the alarm threshold, find out the starting point of degradation, and realize the detection of abnormal points of early degradation of the bearing.…”

Transfer Learning With Time Series Data: A Systematic Mapping Study

“…Applications of fault diagnosis and fault detection are found especially in context of rotating machinery to analyze bearing faults or gearbox faults. Fault detection is addressed in [46] as an anomaly detection problem to distinguish between normal and abnormal states. The other 34 publications address fault diagnosis, which attempts to classify between fault types or health statuses.…”

Logical–Linguistic Model of Diagnostics of Electric Drives with Sensors Support

“…In [ 15 ], the authors presented the machine learning technique as a promising tool for the early detection of rolling bearing failure. To solve the problems of false alarms and thus increase the reliability of the detection results, a robust method of early bearing failure detection based on learning of deep transmission was proposed.…”