[SCIE] SSA-SLTransformer Bearing Fault Diagnosis under Noisy Factory Environments
- 이서영
- 조회수1718
- 2022-03-22
-Title: SSA-SLTransformer Bearing Fault Diagnosis under Noisy Factory Environments
-Journal/Conference: MDPI Electronics Journal (Special Issue: Advances in Fault Dectection/Diagnosis of Electical Power Devices), 11(9), 1504, 7 May 2022
-Authors: Seoyeong Lee and Jongpil Jeong *
-DOI: https://doi.org/10.3390/electronics11091504 / https://www.mdpi.com/2079-9292/11/9/1504
-Journal/Conference Link: https://www.mdpi.com/journal/electronics/special_issues/EPD_electronics
Abstract:
Among the smart factory studies, in this paper, a defect detection study was conducted on bearings, which are elements of mechanical facilities. Bearing research has been consistently conducted in the past. However, most of the research has been limited to using existing artificial intelligence models. In addition, studies that assumed the factory situated in the bearing defect study were insufficient. Therefore, a recent study was conducted assuming the application of the artificial intelligence model and the factory environment. The Transformer model selected as State-of-the-art (SOTA) was also applied to bearing research. Then, an experiment was conducted with Gaussian noise applied to assume the factory situation. The Swish-LSTM Transformer (SLTransformer) framework was constructed by redesigning the internal structure of the Transformer using the Swish activation function and Long Short-Term Memory (LSTM). Then, the noise of the data was removed and reconstructed using the Singular Spectrum Analysis (SSA) preprocessing method. Based on the SSA-SLTransformer framework, an experiment was performed by adding gaussian noise to the Case Western Reserve University (CWRU) dataset. In the case of no noise, SLTransformer showed more than 95\% performance, and when noise was inserted, SSA-SLTransformer showed the best performance than the comparative artificial intelligence models.