Next Article in Journal
The Influence of Boiling on the Streamlined Body Drag Force and Falling Velocity
Next Article in Special Issue
Low-Power Beam-Switching Technique for Power-Efficient Collaborative IoT Edge Devices
Previous Article in Journal
Energy Efficient Secure Communication Model against Cooperative Eavesdropper
Article

A Study on Deep Learning Application of Vibration Data and Visualization of Defects for Predictive Maintenance of Gravity Acceleration Equipment

1
Deparment Electric Computer Engineering, Inha University, 100 Inha-ro, Michuhol-gu, Incheon 22201, Korea
2
Department Energy and Mechanical Engineering, Gyeong-Sang National University, 38, Cheondaegukchi-gil, Tongyeong-si 530-64, Korea
3
R&D Center, ATG, Seongnam-daero, Bundang-gu, Seongnam-si 13558, Korea
4
Department of Otolaryngology-Head and Neck Surgery, Inha Research Institute for Aerospace Medicine, College of Medicine, Inha University, 3-Ga Shinheungdong, Jung-Gu, Incheon 400-711, Korea
*
Author to whom correspondence should be addressed.
Academic Editors: Manuel Armada, Kyungyong Chung and Ellen J. Hong
Appl. Sci. 2021, 11(4), 1564; https://doi.org/10.3390/app11041564
Received: 15 December 2020 / Revised: 28 January 2021 / Accepted: 3 February 2021 / Published: 9 February 2021
Hypergravity accelerators are a type of large machinery used for gravity training or medical research. A failure of such large equipment can be a serious problem in terms of safety or costs. This paper proposes a prediction model that can proactively prevent failures that may occur in a hypergravity accelerator. An experiment was conducted to evaluate the performance of the method proposed in this paper. A 4-channel accelerometer was attached to the bearing housing, which is a rotor, and time-amplitude data were obtained from the measured values by sampling. The method proposed in this paper was trained with transfer learning, a deep learning model that replaced the VGG19 model with a Fully Connected Layer (FCL) and Global Average Pooling (GAP) by converting the vibration signal into a short-time Fourier transform (STFT) or Mel-Frequency Cepstral Coefficients (MFCC) spectrogram and converting the input into a 2D image. As a result, the model proposed in this paper has seven times decreased trainable parameters of VGG19, and it is possible to quantify the severity while looking at the defect areas that cannot be seen with 1D. View Full-Text
Keywords: artificial intelligence; deep learning; fault detection; hyper-gravity machine; vibration monitoring artificial intelligence; deep learning; fault detection; hyper-gravity machine; vibration monitoring
Show Figures

Figure 1

MDPI and ACS Style

Lee, S.; Yu, H.; Yang, H.; Song, I.; Choi, J.; Yang, J.; Lim, G.; Kim, K.-S.; Choi, B.; Kwon, J. A Study on Deep Learning Application of Vibration Data and Visualization of Defects for Predictive Maintenance of Gravity Acceleration Equipment. Appl. Sci. 2021, 11, 1564. https://doi.org/10.3390/app11041564

AMA Style

Lee S, Yu H, Yang H, Song I, Choi J, Yang J, Lim G, Kim K-S, Choi B, Kwon J. A Study on Deep Learning Application of Vibration Data and Visualization of Defects for Predictive Maintenance of Gravity Acceleration Equipment. Applied Sciences. 2021; 11(4):1564. https://doi.org/10.3390/app11041564

Chicago/Turabian Style

Lee, SeonWoo, HyeonTak Yu, HoJun Yang, InSeo Song, JungMu Choi, JaeHeung Yang, GangMin Lim, Kyu-Sung Kim, ByeongKeun Choi, and JangWoo Kwon. 2021. "A Study on Deep Learning Application of Vibration Data and Visualization of Defects for Predictive Maintenance of Gravity Acceleration Equipment" Applied Sciences 11, no. 4: 1564. https://doi.org/10.3390/app11041564

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop