Abstract
The accurate prediction of the remaining useful life (RUL) of rolling bearings is of immense importance in ensuring the safe and smooth operation of machinery and equipment. Although the prediction accuracy has been improved by a predictive model based on deep learning, it is still limited in engineering because lots of models use single-scale features to predict and assume that the degradation data of each bearing has a consistent distribution. In this paper, A deep convolutional migration network based on spatial pyramid pooling (SPP-CNNTL) is proposed to obtain higher prediction accuracy with self-extraction of multi-feature from the original vibrating signal. And to consider the differences of the data distribution in different failure types, transfer learning (TL) added with maximum mean difference (MMD) measurement function is used in the RUL prediction part. Finally, the data of IEEE PHM 2012 Challenge is used for verification, and the results show that the method in this paper has high prediction accuracy
Highlights
As one of the most important components in rotating machinery, rolling bearings play a vital role in the safe operation of mechanical equipment [5]
The domain adaptation module is based on the data distribution difference between the source domain and the target domain in the specified layer, and uses the maximum mean difference (MMD) function value as a measure to constrain the remaining useful life (RUL) prediction part to minimize the difference between the data distribution
This paper proposes a RUL prediction model of bearing based on multi-feature deep convolution transfer learning
Summary
As one of the most important components in rotating machinery, rolling bearings play a vital role in the safe operation of mechanical equipment [5]. This method can theoretically explain the degradation state of machinery, as the complexity of the mechanical system becomes higher and higher, it is difficult to establish an ideal degradation model These statistical model-based methods can achieve predictions under different working conditions, but it is usually assumed that the degraded signal follows a parameterized process model, which may not be the case in reality [33]. Transfer learning based on the MMD function is introduced in the CNN model to solve the problem of low prediction accuracy caused by inconsistent bearing data distribution of different fault types. 2. Transfer learning is used to solve the problem of inconsistent distribution of bearing degradation data and failure data, so as to realize the deep learning model to predict the RUL of different failed bearings.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Eksploatacja i Niezawodność – Maintenance and Reliability
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
