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

Autor: KyuSung Kim, Jungmu Choi, Gang-Min Lim, InSeo Song, Hyeontak Yu, Jang-Woo Kwon, Byeong-Keun Choi, SeonWoo Lee, JaeHeung Yang, HoJun Yang
Jazyk: angličtina
Rok vydání: 2021
Předmět:
0209 industrial biotechnology
vibration monitoring
Computer science
02 engineering and technology
Accelerometer
lcsh:Technology
Fault detection and isolation
Predictive maintenance
law.invention
lcsh:Chemistry
020901 industrial engineering & automation
law
0202 electrical engineering
electronic engineering
information engineering
General Materials Science
Instrumentation
lcsh:QH301-705.5
Simulation
Fluid Flow and Transfer Processes
Rotor (electric)
business.industry
lcsh:T
Process Chemistry and Technology
Deep learning
020208 electrical & electronic engineering
General Engineering
Short-time Fourier transform
deep learning
hyper-gravity machine
artificial intelligence
fault detection
lcsh:QC1-999
Computer Science Applications
Vibration
lcsh:Biology (General)
lcsh:QD1-999
lcsh:TA1-2040
Spectrogram
Artificial intelligence
business
lcsh:Engineering (General). Civil engineering (General)
lcsh:Physics
Zdroj: Applied Sciences, Vol 11, Iss 1564, p 1564 (2021)
Applied Sciences
Volume 11
Issue 4
ISSN: 2076-3417
Popis: 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.
Databáze: OpenAIRE
Externí odkaz: