[Scopus] Anomaly Detection Based on 1D-CNN-LSTM Auto-Encoder for Bearing Data
- 이대희
- 조회수1823
- 2022-11-30
-Title: Anomaly Detection Based on 1D-CNN-LSTM Auto-Encoder for Bearing Data
-Conference: 6th European Conference on Electrical Engineering & Computer Science (ELECS 2022), 21-23 December 2022 (Bern, Switzerland)
-Journal : WSEAS Transactions on Information Science and Applications, Volume 20, Art. #1, p.1-6, 9 January 2023
-Authors: Daehee Lee, Hyunseung Choo, and Jongpil Jeong
-DOI: https://doi.org/10.37394/23209.2023.20.1
-Journal Link: https://wseas.com/journals/isa/index.php
Abstract:
The manufacturing industry is developing rapidly due to the Fourth Industrial Revolution. If a bearing equipment, which is one of the essential parts of the manufacturing industry, fails, it will hinder the production of the manufacturing industry, which will lead to huge losses for the company. To prevent this, this paper implements a 1 Dimension-Convolution Neural Networks-Long Short-Term Memory (1D-CNNLSTM) Auto-Encoder model for fault diagnosis of bearing data. The 1D-CNN-LSTM AutoEncoder model showed high accuracy of 58 to 100 percentage for eccentric bearing data that are difficult to visually diagnose as faults. In the future, we would like to extend this to a real-time failure diagnosis system that can remotely monitor the condition of the bearing equipment through realtime communication with the cloud server and test bed.