Korean Chemical Engineering Research

The Korean Institute of Chemical Engineers (한국화학공학회)
권호 표지
DOI QR코드

DOI QR Code

Fault Detection and Diagnosis for EVA Production Processes Using AE-SOM

AE-SOM을 이용한 EVA 생산 공정 이상 검출 및 진단

  • Received : 2020.03.10
  • Accepted : 2020.04.30
  • Published : 2020.08.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, the AE-SOM method, which combines auto-encoder and self-organizing map, is used to detect and diagnose faults in EVA production process. Then, the fault propagation pathways are identified using Granger causality test. One year and seven months of operation data were obtained to detect faults of the process, and the process variables of the autoclave reactor are mainly analyzed. In the data pretreatment process, the data are standardized and 200 samples of each grade are randomly chosen to obtain a fault detection model. After that, the best matching unit (BMU) of each grade is confirmed by applying AE-SOM. The faults are determined based on each BMU. When a fault is found, the most causative variable of the fault is identified by using a contribution plot, and the fault propagation pathway is identified by Granger causality test. The prognostic of the two shutdowns is detected, and the fault propagation pathway caused by the faulty variable was analyzed.

본 연구에서는 auto-encoder와 self-organizing map을 결합한 auto-encoder with self-organizing map(AE-SOM) 기법을 이용하여 EVA 생산공정의 이상을 검출 및 진단하였고, Granger의 인과분석을 통해 이상 검출 데이터의 이상 전파 방향을 확인하였다. 분석 데이터는 1년 7개월 간의 조업데이터를 이용하였으며, autoclave 반응기의 조업 변수를 주로 분석하였다. 데이터 전처리 과정에서 데이터의 표준화를 먼저 진행하고, 조업의 각 grade의 sample 수를 동일하게 200개 임의로 추출하였다. 이후 AE-SOM을 적용하여 각 grade의 best matching unit (BMU)를 도출하였다. 각각의 BMU를 기준으로 조업 데이터가 얼마나 벗어났는지를 기준으로 데이터의 이상을 판별하였다. 공정 이상이 발견될 시 이상원인을 contribution plot을 이용하여 확인하였고 이상원인 변수의 인과성을 Granger의 인과분석을 통해 분석하였다. 그 결과 조업 시 발생한 2번의 셧다운의 전조를 모두 검출하였으며 이상이 발생한 원인변수에서 기인한 공정 이상의 전파 방향을 분석하였다.

Keywords

References

  1. Sharmin, R., Shah, S. L. and Sundararaj, U., "A PCA Based Fault Detection Scheme for an Industrial High Pressure Polyethylene Reactor," Macromolecular Reaction Engineering, 2(1), 12-30(2008). https://doi.org/10.1002/mren.200700023
  2. Kumar, V., Sundararaj, U., Shah, S. L., Hair, D. and Vande Griend, L. J., "Multivariate Statistical Monitoring of a High-pressure Polymerization Process," Polymer Reaction Engineering, 11, 1017-1052(2003). https://doi.org/10.1081/PRE-120026883
  3. Sivalingam, G., Soni, N. J. and Vakil, S. M., "Detection of Decomposition for High Pressure Ethylene/vinyl Acetate Copolymerization in Autoclave Reactor Using Principal Component Analysis on Heat Balance Model," The Canadian Journal of Chemical Engineering, 93(6), 1063-1075(2015). https://doi.org/10.1002/cjce.22200
  4. Lv, F., Wen, C., Bao, Z. and Liu, M., "Fault Diagnosis Based on Deep Learning," Proceedings of the American Control Conference, 2016 July. Boston. USA, 6851-6856(2016).
  5. Yu, H., Khan, F., Garaniya, V. and Ahmad, A., "Self-organizing Map Based Fault Diagnosis Technique for Non-gaussian Processes," Industrial and Engineering Chemistry Research, 53(21), 8831-8843(2014). https://doi.org/10.1021/ie500815a
  6. Lv, F., Wen, C., Liu, M. and Bao, Z., "Weighted Time Series Fault Diagnosis Based on a Stacked Sparse Autoencoder," Journal of Chemometrics, 31(9), 1-16(2017).
  7. Yu, H., Khan, F. and Garaniya, V., "Risk-based Fault Detection Using Self-organizing Map," Reliability Engineering and System Safety, 139, 82-96(2015). https://doi.org/10.1016/j.ress.2015.02.011
  8. Hinton, G. E. and Salakhutdinov, R. R., "Reducing the Dimensionality of Data with Neural Networks," Science, 313, 504-507 (2006). https://doi.org/10.1126/science.1127647
  9. Bengio, Y., Lamblin, P., Popovici, D. and Larochelle, H., "Greedy Layer-wise Training of Deep Networks," Advances in Neural Information Processing Systems, 153-160(2007).
  10. Tian, J., Azarian, M. H. and Pecht, M., "Anomaly Detection Using Self-organizing Maps-based k-nearest Neighbor Algorithm," In Proceedings of the European Conference of the Prognostics and Health Management Society, 1-9(2014).
  11. Ahmed, U., Ha, D., Shin, S., Shaukat, N., Zahid, U. and Han, C., "Estimation of Disturbance Propagation Path Using Principal Component Analysis (PCA) and Multivariate Granger Causality (MVGC) Techniques," Industrial and Engineering Chemistry Research, 56(25), 7260-7272(2017). https://doi.org/10.1021/acs.iecr.6b02763
  12. Granger, C. W. J., "Investigating Causal Relations by Econometric Models and Cross-spectral Methods," Econometrica: Journal of the Econometric Society, 424-438(1969).