A New Method for Transition Identification by
 Using Dynamic Information

Abstract

Abstract: An identification method is proposed for transitional processes by using dynamic information. A definition for the transitional process is presented here to make a brief description of the characteristics of transition. Because of the existence of dynamic behavior in industrial processes, a dynamic mutual information (DMI) method is proposed in this paper. The differences in statistical characteristics between transition and stable modes can then be well revealed through the DMI algorithm. Extensive tests are carried out both on a numerical example and Tennessee Eastman (TE) benchmark process to validate the effectiveness of the proposed method. The simulation results show the superiority of the proposed method in transition identification and fault detection.

Key words: transition process, multimode transition identification, mutual information, dynamic mutual information

CLC Number:

TP 274

HE Yuchen,GE Zhiqiang,SONG Zhihuan. A New Method for Transition Identification by
Using Dynamic Information[J]. Journal of Shanghai Jiaotong University, 2017, 51(6): 686-692.

References

［1］DAYAL B S, Macgregor J F. Improved PLS algorithms［J］. Journal of Chemometrics, 1998, 11(1): 7385.
［2］GERTLER J, LI W, HUANG Y, et al. Isolation enhanced principal component analysis［J］. Aiche Journal, 1999, 45(2): 323334.
［3］QIN S J, DUNIA R. Determining the number of principal components for best reconstruction［J］. Journal of Process Control, 2000, 10(2/3): 245250.
［4］TRYGG J, WOLD S. Orthogonal projections to latent structures (OPLS)［J］. Journal of Chemometrics, 2002, 16(3): 119128.
［5］ZHOU D, LI G, QIN S J. Total projection to latent structures for process monitoring［J］. Aiche Journal, 2009, 56(1): 168178.
［6］GE Z, SONG Z, GAO F. Review of recent research on databased process monitoring［J］. Industrial & Engineering Chemistry Research, 2013, 52(10): 35433562.
［7］HWANG D H, HAN C. Realtime monitoring for a process with multiple operating modes［J］. Control Engineering Practice, 1999, 7(7): 891902.
［8］BEAVER S, PALAZOGLU A, ROMAGNOLI J A. Cluster analysis for autocorrelated and cyclic chemical process data［J］. Industrial & Engineering Chemistry Research, 2007, 46(11): 36103622.
［9］ZHU Z, SONG Z, PALAZOGLU A. Transition process modeling and monitoring based on dynamic ensemble clustering and multiclass support vector data description［J］. Industrial & Engineering Chemistry Research, 2011, 50(24): 1396913983.
［10］KRASKOV A, Stgbauer H, GRASSBERGER P. Estimating mutual information［J］. Physical Review E Statistical Nonlinear & Soft Matter Physics, 2003, 69(6): 066138.
［11］RUSSELL E L, LEO M S, BRAATZ R D. Datadriven methods for fault detection and diagnosis in chemical processes ［M］. Berlin, Germany: Springer, 2000: 99108.

[1]	WANG Yan, CHEN Yaoran, HAN Zhaolong, ZHOU Dai, BAO Yan. Short-Term Wind Speed Forecasting Model Based on Mutual Information and Recursive Neural Network [J]. Journal of Shanghai Jiao Tong University, 2021, 55(9): 1080-1086.
[2]	ZHAO Chun-yu1* (赵春宇), LIU Jing-jiang1 (刘景江), MA Lun2 (马伦), ZHANG Wei-jun3 (张伟君). Monotonicity Evaluation Method of Monitoring Feature Series Based on Ranking Mutual Information [J]. Journal of shanghai Jiaotong University (Science), 2015, 20(3): 380-384.
[3]	ZHANG Kui* (张葵), ZHANG Xiao-long (张晓龙), XU Xin (徐新), FU Xiao-wei (付晓薇). Mutual Information Optimization Based Dynamic Log-Polar Image Registration [J]. Journal of shanghai Jiaotong University (Science), 2015, 20(1): 61-67.