A Khan, H Hwang, HS Kim - Mathematics, 2021 - mdpi.com
… synthetic dataaugmentation through virtual sensors for the deep learning-based fault diagnosis of a rotating machine with 42 different classes. The original and augmenteddata were …
DH Martins, AA de Lima, MF Pinto, DO Hemerly… - Journal of Intelligent …, 2023 - Springer
… a hybrid method of dataaugmentation to increase the number of … For comparison purposes, four machine learning … is applied for faultdetection and classification in rotating machines. …
… The inputs are different so we consider two ANN structures for the machinefaultdetection application. The first ANN model for the full FFT of the signal has a large structure because the …
T Zhang, C Li, J Chen, S He, Z Zhou - Mechanical Systems and Signal …, 2023 - Elsevier
… To address this problem, this paper proposes a feature-level consistency regularized semi-supervised scheme with dataaugmentation for faultdiagnosis of machines. In the proposed …
K Yu, TR Lin, H Ma, X Li, X Li - Mechanical Systems and Signal Processing, 2021 - Elsevier
… dataaugmentation (DA) and metric learning is proposed for an intelligent bearing faultdiagnosis under limited labeled data… on an experimental bearing fault dataset from our laboratory …
… accuracy of faultdiagnosis. Moreover, … faultdiagnosis, which is the basis for building a new model. Secondly, we introduce the GAN dataaugmentation, and the generated training data …
X Gao, F Deng, X Yue - Neurocomputing, 2020 - Elsevier
… built to generate auxiliary data for the low-data original dataset in industrial process for fault diagnosis. Since the focus of this paper is the ability of GAN based dataaugmentation, the …
J Wang, B Han, H Bao, M Wang… - Proceedings of the …, 2020 - journals.sagepub.com
… Therefore, the conditional GAN (CGAN) 19 is first used in this paper to generate vibration signals for dataaugmentation of all the fault types in one time. In addition, spectrum …
Q Guo, Y Li, Y Liu, S Gao, Y Song - IEEE Sensors Journal, 2022 - ieeexplore.ieee.org
… samples for imbalanced faultdata with normal data and less faulty data. The experiment … [21] proposed a novel faultdiagnosis model based on dataaugmentation techniques called …