Abstract
In industry there are many complex modeling tasks where the most of the available information is in the form of input-output data. In such cases only black box modeling can be used, where the model can be built using learning methods. In black-box modeling one of the most important tasks is to obtain good training data. However, in most real world problems the available data are imprecise, contain noise or some distortion. This paper discusses some problems of neural model building based on noisy training data. Two methods - the er- rors-in-variables training method (EIV) and the support vector machines (SVM)- are introduced and compared to the performance of the traditional neural network solution. The performance of the SVM method is also tested on a real industrial problem, namely on the modeling of a Linz-Donawitz steel converter.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Andersson, B. D. O.: “Identification of scalar errors in variables models with dynamics” Automatica, Vol. 21. No. 6. pp. 709–716., 1985
Van Gorp, J., Schoukens, J. and Pintelon, R.: “Learning neural networks with noisy inputs using the errors-in-variables approach” IEEE Trans. on Neural Networks, Vol. 11. No. 2 pp. 402–414. 2000.
Vapnik, V. N. “Statistical Learning Theory”, John Wiley, 1998.
Wang, Ch. and Principe, J. C. “Training neural networks with additive noise in the desired signal” IEEE Trans. on Neural Networks, Vol. 10. No. 6., pp. 1511–1517. 1999.
An, G. “The effect of adding noise during backpropagation training on the generalization performance” Neural Computation, Vol. 8, pp. 643–674. 1996.
Gunn, S.: “Support Vector Machines for Classification and Regression”, ISIS Technical Report, 14 May 1998.
Haykin, S.: “Neural Networks, A Comprehensive Foundation” Prentice Hall, New Jersey, 1999.
Smola, A. and Schölkopf, B.: “A Tutorial on Support Vector Regression” NeuroCOLT2 Technical Report Series NC2-TR-1998-030, October, 1998
Joachims, T.: “Making large-Scale SVM Learning Practical. Advances in Kernel Methods-Support Vector Learning”, in Schölkopf, B., Burges, C. and Smola, A. (eds.), MITPress, 1999
Platt, J. C.: “Sequential Minimal Optimization: Fast Algorithm for Training Support Vector Machines” Microsoft Research Technical Report MSR-TR-98-14, April 21, 1998
Polkovnyikov, A.: “Possibilities of Modeling of an LD Converter”, Inner report (In Hungarian) Dunaújváros, 1996.
Kaptay, Gy. and Benkö M.: “The physical-chemical Backgroud of Modeling of an LD Converter”, Inner report, (In Hungarian) Miskolc University, Miskolc, 1999.
Horváth, G., Pataki, B. and Strausz Gy.: “Black-box Modeling of a Complex Industrial Process”, Proc. of the 1999 IEEE Conference and Workshop on Engineering of Computer Based Systems, Nashville, TN, USA. 1999. pp. 60–66
Pataki, B., Horváth, G., Strausz, Gy. and Talata, Zs. “Inverse Neural Modeling of a Linz-Donawitz Steel Converter” e & i Elektrotechnik und Informationstechnik, Vol. 117. No. 1.. pp. 13–17. 2000.
Horváth, G., Pataki, B. and Strausz Gy.: “Research Report of the Hybrid-neural Modeling of an LD Steel-Converter” Budapest University of Technology and Economics, Dept. of Measurement and Information Systems, (In Hungarian) Budapest, 2000.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Berényi, P., Valyon, J., Horváth, G. (2001). Neural Modeling of an Industrial Process with Noisy Data. In: Monostori, L., Váncza, J., Ali, M. (eds) Engineering of Intelligent Systems. IEA/AIE 2001. Lecture Notes in Computer Science(), vol 2070. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45517-5_31
Download citation
DOI: https://doi.org/10.1007/3-540-45517-5_31
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-42219-8
Online ISBN: 978-3-540-45517-2
eBook Packages: Springer Book Archive