Abstract
Recent convolutional neural network (CNN) models in image processing can be used as feature-extraction methods to achieve high accuracy as well as automatic processing in bearing fault diagnosis. The combination of deep learning methods with appropriate signal representation techniques has proven its efficiency compared with traditional algorithms. Vital electrical machines require a strict monitoring system, and the accuracy of these machines’ monitoring systems takes precedence over any other factors. In this paper, we propose a new method for diagnosing bearing faults under variable shaft speeds using acoustic emission (AE) signals. Our proposed method predicts not only bearing fault types but also the degradation level of bearings. In the proposed technique, AE signals acquired from bearings are represented by spectrograms to obtain as much information as possible in the time–frequency domain. Feature extraction and classification processes are performed by deep learning using EfficientNet and a stochastic line-search optimizer. According to our various experiments, the proposed method can provide high accuracy and robustness under noisy environments compared with existing AE-based bearing fault diagnosis methods.
Highlights
High-power, heavy, and large-size industrial electric motors play a vital role in the production process at factories
Optimizing Short-time Fourier Transform (STFT) depends on (1) finding compatible window size; (2) overlap between segments results are obtained for compound faults (inner and outer raceway cracks (BCIO), outer and roller, which affects the density in time; (3) zero-padding for Fast Fourier Transforms (FFTs) calculation; and (4) choosing cracks (BCOR), inner and roller cracks (BCIR), inner–outer–roller cracks (BCIOR)) and normal a suitable signal segment size
In each of Dataset 1 the acoustic emission (AE) signals collected at shaft speeds of 300, 400, and 500 rpm, whereas the validation and testing and Dataset 2, the total number of samples used for the training process
Summary
High-power, heavy, and large-size industrial electric motors play a vital role in the production process at factories. In common fault-diagnosis (FD) processes, the signals which contain the machine health status are acquired and transferred to the server system in which they are analyzed to diagnose faults This approach is expensive compared with the use of handheld FD instruments using the diagnosis method for bearing faults [1], it can improve reliability, which is undeniable in machine monitoring, especially for essential machines. Recent methods for compound bearing fault diagnosis under variable speeds used generic CNN architecture with three convolution layers based on Lenet-5 and enhanced the CNN training process by a stochastic diagonal Levenberg–Marquardt algorithm [10,11]. In these methods, AE signals are converted into two-dimensional spectral energy distribution maps to feed the CNN. For applying EfficientNet to a fault classification process effectively, the optimizer named stochastic line search (SLS) [13] is adopted in the training stage, resulting in improved convergence ability, decreased training time, and reduced number of training samples
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.
