이상치 데이터를 고려한 DT-CNN 기반의 전동기 고장 예측

다국어 초록 (Multilingual Abstract)

One of the major problems with the existing motor failure prediction system is to assume that all motors with the same fault condition have the same or a similar signal. This is a problem that arises because it is impossible to measure all the countless types of motors and data of driving conditions and failures. It is difficult to implement a general-purpose failure prediction system with an existing system having limited data and limited output. Data that have a large difference because they do not exist in the existing system are called outlier data. In previous studies, the problem arising from the outlier data has not been considered. To solve this problem, a system designed by separating the failure diagnosis model and the failure prediction model is proposed. The diagnostic model of the proposed system can detect data that are not inside big data using a decision-tree convolution neural network (DT-CNN). By using the diagnostic model and the predictive model in series, it is possible to analyze data in a non-measured state more efficiently. Additionally, a method for averaging the outputs of the diagnostic and predictive models is proposed. Through this, the deep learning algorithm can obtain in effect of applying the filter. Furthermore, the average values can be used to confirm the long-term signal change trend. The proposed system improves the problems of the existing failure prediction and enables more practical failure prediction.

목차 (Table of Contents)

• Abstract
• I. 서론
• II. 고장 진단 모델
• III. 고장 예측 모델
• IV. 실험
• V. 결론
• REFERENCES

참고문헌 (Reference)

1 성상만, "비선형 측정식을 위한 불변 확장칼만필터 설계 및 이동 로봇에의 적용" 제어·로봇·시스템학회 25 (25): 512-518, 2019

2 임대의, "두 개의 컨볼루션 신경망을 이용한 PCB 상의 SMD 분류 시스템" 제어·로봇·시스템학회 25 (25): 923-928, 2019

3 B. Saied, "fault prediction of deep bar cage rotor induction motor based on FEM" 53 : 2013-, 2013

4 W. J. Scheirer, "Toward open set recognition" 35 (35): 1757-1772, 2013

5 E. Germen, "Sound based induction motor fault diagnosis using Kohonen self-organizing map" 46 : 45-48, 2014

6 W. J. Scheirer, "Probability models for open set recognition" 36 (36): 23172324-, 2014

7 Q. Xu, "Overfitting remedy by sparsifying regularization on fully-connected layers of CNNs" 328 : 69-74, 2018

8 N. Mehala, "Motor current signature analysis and its applications in induction motor fault diagnosis" 2 (2): 29-35, 2007

9 한지훈, "Hyperparameter Optimization Using a Genetic Algorithm Considering Verifcation Time in a Convolutional Neural Network" 대한전기학회 15 (15): 721-726, 2020

10 G. Niu, "Decision-level fusion based on wavelet decomposition for induction motor fault diagnosis using transient current signal" 35 (35): 918-928, 2008

11 A. Li, "Coupled bias-variance trade off for cross pose face recognition" 2011

12 J. H. Han, "A study on motor poor maintenance detection based on DT-CNN" 1234-1237, 2019

13 T. Petsche, "A neural network autoassociator for induction motor failure prediction" 8 : 924-930, 1996