# Combined Input Training and Radial Basis Function Neural Networks based Nonlinear Principal Components Analysis Model Applied for Process Monitoring

### Messaoud Bouakkaz, Mohamed-Faouzi Harkat

#### Abstract

In this paper a novel Nonlinear Principal Component Analysis (NLPCA) is proposed. Generally, a NLPCA model is performed by using two sub-models, mapping and demapping. The proposed NLPCA model consists of two cascade three-layer neural networks for mapping and demapping, respectively. The mapping model is identified by using a Radial Basis Function (RBF) neural networks and the demapping is performed by using an Input Training neural networks (IT-Net). The nonlinear principal components, which represents the desired output of the first network, are obtained by the IT-NET. The proposed approach is illustrated by a simulation example and then applied for fault detection and isolation of the TECP process.

#### References

- Dong, D. and McAvoy, T. (1996). Nonlinear principal component analysis based on principal curves and neural networks. Computers and Chemical Engineering 20, 65 78.
- Downs, J. and Vogel, E. (1993). A plant-wide industrial control problem. Computers and chemical engineering Journal 17, 245-255.
- Dunia, R., Qin, S., Ragot, J., and McAvoy, T. (1996). Identification of faulty sensors using principal component analysis. AIChE Journal 42, 2797-2812.
- Harkat, M., Djellel, S., Doghmane, N., and Benouareth, M. (2007). Sensor fault detection, isolation and reconstruction using nonlinear principal component analysis. Intarnational Journal of Automation and Computing, 4,.
- Harkat, M., Mourot, G., and Ragot, J. (2003). Variable reconstruction using rbf-nlpca for process monitoring. In IFAC Symposium on Fault Detection, Supervision and Safety for Technical Process, SAFEPROCESS. Washington, USA.
- Hastie, T. and Stuetzle, W. (1989). Principal curves. Journal of the American Statistical Association 84, 502-516.
- Hsieh, W. and Li, C. (2001). Nonlinear principal component analysis by neural networks. Tellus Journal 53A, 599- 615.
- Kramer, M. (1991). Nonlinear principal component analysis using auto-associative neural networks. AIChE Journal 37, 233-243.
- LeBlanc, M. and Tibshirani, R. (1994). Adaptive principal surfaces. Journal of American Statistical Association 89(425), 53-64.
- Tan, S. and Mavrovouniotis, M. (1995). Reduction data dimensionality through optimizing neural network inputs. AIChE Journal 41, 1471-1480.
- Verbeek, J. (2001). A k-segments algorithm for finding principal curves. IAS Technical Journal.
- Vogel, N. R. E. (1994). Optimal steady-state operation of the tennessee eastman challenge process. Computers and chemical engineering Journal 19, 949-959.
- Webb, A., Vlassis, N., and Krose, B. (1999). A loss function to model selection in nonlinear principal components. Neural Networks Journal 12, 339-345.
- Zhu, Q. and Li, C. (2006). Dimensionality reduction with input training neural network and its application in chemical process modeling. Chinese Journal.

#### Paper Citation

#### in Harvard Style

Bouakkaz M. and Harkat M. (2012). **Combined Input Training and Radial Basis Function Neural Networks based Nonlinear Principal Components Analysis Model Applied for Process Monitoring** . In *Proceedings of the 4th International Joint Conference on Computational Intelligence - Volume 1: NCTA, (IJCCI 2012)* ISBN 978-989-8565-33-4, pages 483-492. DOI: 10.5220/0004152304830492

#### in Bibtex Style

@conference{ncta12,

author={Messaoud Bouakkaz and Mohamed-Faouzi Harkat},

title={Combined Input Training and Radial Basis Function Neural Networks based Nonlinear Principal Components Analysis Model Applied for Process Monitoring},

booktitle={Proceedings of the 4th International Joint Conference on Computational Intelligence - Volume 1: NCTA, (IJCCI 2012)},

year={2012},

pages={483-492},

publisher={SciTePress},

organization={INSTICC},

doi={10.5220/0004152304830492},

isbn={978-989-8565-33-4},

}

#### in EndNote Style

TY - CONF

JO - Proceedings of the 4th International Joint Conference on Computational Intelligence - Volume 1: NCTA, (IJCCI 2012)

TI - Combined Input Training and Radial Basis Function Neural Networks based Nonlinear Principal Components Analysis Model Applied for Process Monitoring

SN - 978-989-8565-33-4

AU - Bouakkaz M.

AU - Harkat M.

PY - 2012

SP - 483

EP - 492

DO - 10.5220/0004152304830492