스펙트로그램 이미지를 이용한 CNN 기반 자동화 기계 고장 진단 기법 = CNN-based Automatic Machine Fault Diagnosis Method Using Spectrogram Images

국문 초록 (Abstract)

소리 기반 기계 고장 진단은 기계의 음향 방출 신호에서 비정상적인 소리를 자동으로 감지하는 것이다. 수학적 모델을 사용하는 기존의 방법은 기계 시스템의 복잡성과 잡음과 같은 비선형 요인이 존재하기 때문에 기계 고장 진단이 어려웠다. 따라서 기계 고장 진단의 문제를 딥러닝 기반 이미지 분류 문제로 해결하고자 한다. 본 논문에서 스펙트로그램 이미지를 이용한 CNN 기반 자동화 기계 고장 진단 기법을 제안한다. 제안한 방법은 기계의 결함 시 발생하는 주파수상의 특징 벡터를 효과적으로 추출하기 위해 STFT를 사용하였으며, STFT에 의해 검출된 특징 벡터들은 스펙트로그램 이미지로 변환하여 CNN을 이용해 기계의 상태별로 분류한다. 그 결과는 제안한 방법은 효과적으로 결함을 탐지할 뿐만 아니라 소리 기반의 다양한 자동 진단 시스템에도 효과적으로 활용될 수 있다.

다국어 초록 (Multilingual Abstract)

Sound-based machine fault diagnosis is the automatic detection of abnormal sound in the acoustic emission signals of the machines. Conventional methods of using mathematical models were difficult to diagnose machine failure due to the complexity of the industry machinery system and the existence of nonlinear factors such as noises. Therefore, we want to solve the problem of machine fault diagnosis as a deep learning-based image classification problem. In the paper, we propose a CNN-based automatic machine fault diagnosis method using Spectrogram images. The proposed method uses STFT to effectively extract feature vectors from frequencies generated by machine defects, and the feature vectors detected by STFT were converted into spectrogram images and classified by CNN by machine status. The results show that the proposed method can be effectively used not only to detect defects but also to various automatic diagnosis system based on sound.

참고문헌 (Reference)

1 B. Robert, "Rolling element bearing diagnostics" 25 (25): 485-520, 2011

2 T. Ince, "Real-time motor fault detection by 1D convolutional neural networks" 63 (63): 7067-7075, 2016

3 C. Chen, "Machine condition prediction based on adaptive neuro-fuzzy and high-order particle filtering" 58 (58): 4353-4364, 2011

4 J. Jiao, "Hierarchical discriminating sparse coding for weak fault feature extraction of rolling bearings" 184 : 41-54, 2019

5 B. Sreejith, "Fault diagnosis of rolling element bearing using time-domain features and neural networks" 1-6, 2009

6 P. K. Kankar, "Fault diagnosis of ball bearings using machine learning methods" 38 (38): 1876-1886, 2011

7 J. D. Zheng, "Enhanced empirical wavelet transform based time-frequency analysis and its application to rolling bearing fault diagnosis" 46 (46): 358-364, 2018

8 W. Sun, "Convolutional discriminative feature learning for induction motor fault diagnosis" 13 (13): 1350-1359, 2017

9 M. Gan, "Construction of hierarchical diagnosis network based on deep learning and its application in te fault pattern recognition of rolling element bearings" 72-73 (72-73): 7067-7075,

10 T. H. loutas, "Condition monitoring of a single stage gearbox with artificially induced gear cracks utilizing on-line vibration and acoustic emission measurements" 70 (70): 1148-1159, 2009

11 이경민, "Automatic Machine Fault Diagnosis System using Discrete Wavelet Transform and Machine Learning" 한국멀티미디어학회 20 (20): 1299-1311, 2017

12 Y. Lei, "An intelligent fault diagnosis method using unsupervised feature learning towards mechanical big data" 63 (63): 3137-3147, 2016

13 Y. S. Wang, "An intelligent approach for engine fault diagnosis based on hibert-huang transform and support vector machine" 75 : 1-9, 2014

14 D. Z. Li, "An enhanced bispectrum technique with auxiliary frequency injection for induction motor health condition monitoring" 64 (64): 2679-2687, 2015

15 M. Zhao, "A novel strategy for signal denoising using reweighted SVD and its applications to weak fault feature enhancement of rotating machinery" 94 : 129-147, 2017

16 J. Wan, "A manufacturing big data solution for active preventive maintenance" 13 (13): 2039-2047, 2017