Soft sensor for change detection

Soft sensor for change detection

José Ragot Benoît Marx Didier Maquin 

Université de Lorraine, CRAN, UMR 7039,

2 avenue de la forêt de Haye, 54156 Vandoeuvre-lès-Nancy, France

CNRS, CRAN, UMR 7039, France

Corresponding Author Email: 
jose.ragot@univ-lorraine.fr, benoit.marx@univ-lorraine.fr, didier.maquin@univ-lorraine.fr
Page: 
95-115
|
DOI: 
https://doi.org/10.3166/I2M.15.3-4.95-115
Received: 
N/A
| |
Accepted: 
N/A
| | Citation
Abstract: 

Regime change detection is concerned with identifying abnormal system behaviors and abrupt changes from one regime to another. This paper proposes a novel method capable of detecting regime change points in sequential time series. Our approach is based on a sensitivity study of a global model combining, with a multiplicative effect, the local models describing the different modes of functioning.

Keywords: 

soft sensor, mode change detection, parameter estimation.

1. Introduction
2. Principe de reconnaissance de mode de fonctionnement
3. Système SIMO à deux modes de fonctionnement
4. Influence des erreurs de mesure
5. Généralisation
6. Exemple numérique
7. Conclusion
  References

Adar S., Shkolnisky Y., Ben-Dor E. (2014). Change detection of soils under small-scale laboratory conditions using imaging spectroscopy sensors. Geoderma, vol. 216, p. 19-29. Consulté sur http://www.sciencedirect.com/science/article/pii/S0016706113003704

Azimi-Sadjadi B., Krishnaprasadz P. (2004). Change detection for nonlinear systems; a particle filtering approach. EURASIP Journal on Applied Signal Processing, vol. 15, p. 2295-2305.

Bazart L., Maquin D., Khelassi A., Bèle B., Ragot J. (2013, octobre). Operating mode recognition: Application in continuous casting. In 2nd International Conference on Control and Fault-Tolerant Systems, SysTol’13, p. CDROM. Nice, France.

Belkoura L. (2009, mai). Change point detection with application to the identification of a switching process. In International conference on systems theory : Modelling, Analysis and Control. Fes, Morocco. Consulté sur https://hal.inria.fr/inria-00363679

Benaicha A., Mourot G., Benothman K., Ragot J. (2013). Determination of principal component analysis models for sensor fault detection and isolation. International Journal of Control, Automation and Systems, vol. 11, no 2, p. 296–305.

Benaicha A., Mourot G., Guerfel M., Benothman K., Ragot J. (2010, juin). A new method for determining PCA models for system diagnosis. In 18th Mediterranean Conference on Control and Automation, MED’10, p. CDROM. Marrakech, Morocco.

Bildirici M., Ersin O. (2014). Modeling markov switching ARMA-GARCH neural networks models and an application to forecasting stock returns. The Scientific World Journal, vol. 2014, p. 1-21.

Celtin M., Comert G. (2006). Short-term traffic flow prediction with regime switching models. Journal of the Transportation Research Board, vol. 1965, p. 23-31.

Chetouani Y. (2011). Change detection in a distillation column based on the generalized likelihood ratio approach. Journal of Chemical Engineering and Process Technology, vol. 2, no 5, p. 1-6.

Chib S. (1998). Estimation and comparison of multiple change-point models. Journal of Econometrics, vol. 86, no 2, p. 221 - 241.

Chithra N., Thampi S. G. (2015). Detection and attribution of climate change signals in precipitation in the chaliyar river basin, kerala, india. Aquatic Procedia, vol. 4, p. 755 - 763.

Domlan E., Huang B., Ragot J., Maquin D. (2009, December). Robust identification of switched regression models. IET Control Theory and Applications, vol. 3, no 12, p. 1578-1590.

Gustafsson F. (2000). Adaptive filtering and change detection (J. Wiley, Ed.).

Hayton P., Utete S., King D., King S., Anuzis P., Tarassenko L. (2007). Static and dynamic novelty detection methods for jet engine health monitoring. Philosophical Transactions of the Royal Society, vol. 1851, p. 493-514.

Hinkley D. (1971). Inference about the change point from cumulative sum tests. Biometrika, vol. 58, p. 509-523.

Hub J., Groot B. de. (2009). Detection of functional modes in protein dynamics. PLoS Computational Biology, vol. 5, no 8.

Hwang I., Balakrishnan H., Tomlin C. (2007). State estimation for hybrid systems: applications to aircraft tracking. IEE Proceedings of Control Theory and Applications, vol. 153, no 5, p. 556-566.

Jolliffe I. (2005). Principal component analysis. Springer.

Kallas M., Mourot G., Maquin D., Ragot J. (2015). Fault estimation of nonlinear processes using kernel principal component analysis. In 14th european control conference. Linz, Austria.

Koyama K.,Watanabe K., Kobayashi K., Kurihara Y. (2006). A multi-variable detecting sensor and its application. In Sice-icase international joint conference.

Laurent H., Doncarli C. (1996). Abrupt changes detection in the time-frequency plane. In Ieee-sp international symposium on time-frequency and time- scale analysis, p. 285-288. Paris, France.

Mantua N. (2004). Methods for detecting regime shifts in large marine ecosystems: a review with approaches applied to north pacific data. Progress in Oceanography, vol. 60, p. 165-182.

Martínez-Beneito M., Conesa D., López-Quílez A., López-Maside A. (2008). Bayesian markov switching models for the early detection of influenza epidemic. Statistics in Medicine, vol. 27, no 22.

Mc Elroya L., Baoa J., Yangb R., Yub A. (2008). A soft-sensor approach to flow regime detection for milling processes. Powder Technology, vol. 188, no 3, p. 234-241.

Nielsen A. A., Canty M. J. (2009). Kernel principal component and maximum autocorrelation factor analyses for change detection. In Society of photo-optical instrumentation engineers (spie) conference series, vol. 7477.

Pekpe K., Mourot G., Gasso K., Ragot J. (2004). Identification of switching systems using change detection technique in the subspace framework. In Conference on decision and control, vol. 4, p. 3720-3725.

Ragot J., Darouach M., Maquin D., Bloch G. (1990). Validation de données par équilibrage de bilan. Hermès, Paris, France.

Ragot J., Hocine A., Maquin D. (2004). Parameter estimation of switching systems. In International conference on computational intelligence for modelling, control and automation,. Gold Coast, Australie.

Rigatos G., Rigatou E. (2015). Change detection in the dynamics of an intracellular protein synthesis model using nonlinear Kalman filtering. 9th IFAC Symposium on Biological and Medical Systems, vol. 48, no 20, p. 267 - 272. Consulté sur http://www.sciencedirect.com/science/article/pii/S2405896315020418

Tarantino C., Adamo M., Lucas P., R. Blonda. (2016). Detection of changes in semi-natural grasslands by cross correlation analysis with worldview-2 images and new landsat 8 data. Remote Sensing of Environment,, vol. 175, p. 65-72.

Zhu X., Xie Y., Li J., Wu D. (2015). Change point detection for subprime crisis in american banking: From the perspective of risk dependence. International Review of Economics & Finance, vol. 8, p. 18-28.