DOI QR코드

DOI QR Code

오토인코더를 이용한 열간 조압연설비 상태모니터링과 진단

Condition Monitoring and Diagnosis of a Hot Strip Roughing Mill Using an Autoencoder

  • 투고 : 2019.02.07
  • 심사 : 2019.02.25
  • 발행 : 2019.03.31

초록

Purpose: It is essential for the steel industry to produce steel products without unexpected downtime to reduce costs and produce high quality products. A hot strip rolling mill consists of many mechanical and electrical units. In condition monitoring and diagnosis, various units could fail for unknown reasons. Methods: In this study, we propose an effective method to detect units with abnormal status early to minimize system downtime. The early warning problem with various units was first defined. An autoencoder was modeled to detect abnormal states. An application of the proposed method was also implemented in a simulated field-data analysis. Results: We can compare images of original data and reconstructed images, as well as visually identify differences between original and reconstruction images. We confirmed that normal and abnormal states can be distinguished by reconstruction error of autoencoder. Experimental results show the possibility of prediction due to the increase of reconstruction error from just before equipment failure. Conclusion: In this paper, hot strip roughing mill monitoring method using autoencoder is proposed and experiments are performed to study the benefit of the autoencoder.

키워드

PJGOB9_2019_v47n1_75_f0002.png 이미지

Figure 1. A Schematic illustration of hot strip rolling mill process

PJGOB9_2019_v47n1_75_f0003.png 이미지

Figure 2. Illustration and image(left) and picture of a roughing mill(right)

PJGOB9_2019_v47n1_75_f0004.png 이미지

Figure 3. Drive and non-drive side vibration(Top : Drive side, Bottom : Non Drive side)

PJGOB9_2019_v47n1_75_f0005.png 이미지

Figure 4. Scatter plots of evaluation data set(Normal: solid dots, Abnormal: empty dots)

PJGOB9_2019_v47n1_75_f0006.png 이미지

Figure 5. Data structures

PJGOB9_2019_v47n1_75_f0007.png 이미지

Figure 6. Autoencoder(right) and stacked autoencoder(left) structures

PJGOB9_2019_v47n1_75_f0008.png 이미지

Figure 7. Reconstruction error trend of autoencoder model

PJGOB9_2019_v47n1_75_f0009.png 이미지

Figure 8. Reconstruction error plot

PJGOB9_2019_v47n1_75_f0010.png 이미지

Figure 9. Original and reconstruction images

Table 1. Hyper-parameter for autoencoder modeling

PJGOB9_2019_v47n1_75_t0001.png 이미지

Table 2. Stacked autoencoder modeling

PJGOB9_2019_v47n1_75_t0002.png 이미지

참고문헌

  1. Ahmad, S., Lavin, A., Purdy, S., and Agha, Z. 2017. "Unsupervised real-time anomaly detection for streaming data." Neurocomputing 262(1):134-147. https://doi.org/10.1016/j.neucom.2017.04.070
  2. Choi. J. H. 2013. "Introduction of Prognostics and Health Management." Journal of the Korean Society of Mechanical Engineers 53(7):26-34.
  3. Devika, U.K., and Sunny, D. 2014. "Comparison Study among Various Anomaly Detection Techniques." Journal of Research in Computer and Communication Technology 3(3):121-124.
  4. Farina, M., Osto. E., Perizzato, A., Piroddi, L., and Scattolini, R. 2015. "Fault Detection and Isolation of Bearings in a Drive Reducer of a Hot Steel Rolling Mill." Control Engineering Practice 39:35-44. https://doi.org/10.1016/j.conengprac.2015.02.001
  5. Hansson, K., Yella, S., Dougherty, M., and Fleyeh, H. 2016. "Machine Learning Algorithms in Heavy Process Manufacturing." American Journal of Intelligent Systems 6(1):1-13.
  6. Hu, Y., Palme, T., and Fink, O. 2016. "Deep Health Indicator Extraction: A Method Based on Auto-encoders and Extreme Learning Machines." Annual Conference of the Prognostics and Health Management Society:1-7.
  7. Kumar, S., Chow, W. S., and Pecht, M. 2010. "Approach to Fault Identification for Electronic Products Using Mahalanobis Distance." IEEE Transactions on Instrumentation and Measurement 59(8):20552064.
  8. Lee, S. H., and Yun, B. D. 2015. "Industry 4.0 and Direction of Failure Prediction and Health Management Technology(PHM)." Journal of the Korean Society for Noise and Vibration Engineering 25(1):22-28.
  9. Omar, S., Ngadi, A., and Jebur, H. H. 2013. "Machine Learning Techniques for Anomaly Detection : An overview." International Journal of Computer Applications 79(2):33-41. https://doi.org/10.5120/13715-1478
  10. Sarkar, S., Reddy, K. K., Giering, M., and Gurvich, M. R. 2016. "Deep Learning for Structural Health Monitoring: A damage characterization application." Annual Conference of the Prognostics and Health Management Society:176-182.
  11. Seo, M. K., and Yun, W. Y. 2017. "Clustering-based Monitoring and Fault detection in Hot Strip Roughing Mill." Journal of the Korean Society for Quality Management 45(1):25-37. https://doi.org/10.7469/JKSQM.2017.45.1.025
  12. Verdier, G., and Ferreira, A. 2011. "Adaptive Mahalanobis Distance and k-Nearest Neighbor Rule for Fault Detection in Semiconductor Manufacturing." IEEE Transactions Semiconductor Manufacturing 24(1):59-68. https://doi.org/10.1109/TSM.2010.2065531
  13. Vincent, P., Larochelle, H., Lajoie, I., Bengio, Y., and Manzagol, P. A. 2010. "Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion." The Journal of Machine Learning Research 11(3):3371-3408.
  14. Yacout, S. 2010. "Fault Detection and Diagnosis for Condition Based Maintenance Using the Logical Analysis of Data." Computers and Industrial Engineering(CIE) 2010 40th International Conference:1-6.
  15. Zhou, Y., Li, H.M., Huang, D., and Gao, Z. 2015. "Data Driven Injection Molding Process Monitoring Using Sparse Autoencoder Technique." IEEE International Conference on Advanced Intelligent Mechatronics(AIM):524-528.
  16. Zhu, H. J., Ting, R., Wang, X. Q., You, Z., and Fang, H. 2015. "Fault Diagnosis of Hydraulic Pump Based on Stacked Autoencoders." IEEE 12th International Conference on Electronic Measurement & Instruments:58-62.