Abstract
As key equipment in modern industry, it is important to diagnose and predict the health status of bearings. Data-driven methods for remaining useful life (RUL) prognostics have achieved excellent performance in recent years compared to traditional methods based on physical models. In this paper, we propose a novel data-driven method for predicting the remaining useful life of bearings based on a deep graph convolutional neural network with spatiotemporal domain convolution. This network uses the average sliding root mean square (ASRMS) as the health factor to identify the healthy and degraded states, and then uses correlation coefficient analysis on the hybrid features of the degraded data to construct a spatial graph according to the strength of the correlation between the obtained features. In the time domain, we introduce historical data as the input to the temporal convolution. After the data are processed by the spatial map and the temporal dimension, we perform the prediction of the remaining useful life. The experimental results show the accuracy of the method.
Highlights
Prognostic and health management (PHM) is a critical technology to ensure the safety and reliability of equipment, which has achieved fruitful theoretical results in the past decades and has been widely applied [1,2,3,4,5]
We find that the curve of average sliding root mean square (ASRMS) is smoother than that of root mean square (RMS), and it can well avoid the interference of outlier points to the judgment of the first degradation point (FDP)
The information contained in the time sequences can have a considerable impact on the prediction of remaining useful life (RUL), so the feature extraction on the temporal dimension of the time sequences signals is performed using temporal convolution network (TCN)
Summary
Prognostic and health management (PHM) is a critical technology to ensure the safety and reliability of equipment, which has achieved fruitful theoretical results in the past decades and has been widely applied [1,2,3,4,5]. Pan et al [24] proposed a two-step prediction method based on ELMs to identify the healthy and degraded states by constructing HI curves. Liang et al [25] extracted hybrid features from the data and used recurrent neural network (RNN) to construct health indicter (HI) curves for the extracted features, which achieved good prediction results. A new technique developed in recent years, provides a new approach for training large amounts of data with its powerful feature extraction capability. Proposed a deep belief network (DBN)-based modeling of the bearing degradation process using a particle swarm algorithm for optimal parameter search, showing a more powerful capability than the traditional RBF. Compared with discrete CNNs, graph neural networks are able to extract deep-level features for high-dimensional data without destroying the topology of data.
Sign-in/Register to view highlights & summary 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 callDisclaimer: 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.
