Semi-mechanistic models for state-estimation - Soft sensor for polymer melt index prediction

Balazs Feil, J. Abonyi, Peter Pach, Sandor Nemeth, Peter Arva, Miklos Nemeth, Gabor Nagy

Research output: Conference contribution

14 Citations (Scopus)

Abstract

Nonlinear state estimation is a useful approach to the monitoring of industrial (polymerization) processes. This paper investigates how this approach can be followed to the development of a soft sensor of the product quality (melt index). The bottleneck of the successful application of advanced state estimation algorithms is the identification of models that can accurately describe the process. This paper presents a semi-mechanistic modeling approach where neural networks describe the unknown phenomena of the system that cannot be formulated by prior knowledge based differential equations. Since in the presented semi-mechanistic model structure the neural network is a part of a nonlinear algebraic-differential equation set, there are no available direct input-output data to train the weights of the network. To handle this problem in this paper a simple, yet practically useful spline-smoothing based technique has been used. The results show that the developed semi-mechanistic model can be efficiently used for on-line state estimation.

Original languageEnglish
Title of host publicationLecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)
EditorsL. Rutkowski, J. Siekmann, R. Tadeusiewicz, L.A. Zadeh
Pages1111-1117
Number of pages7
Volume3070
Publication statusPublished - 2004
Event7th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2004 - Zakopane, Poland
Duration: jún. 7 2004jún. 11 2004

Other

Other7th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2004
CountryPoland
CityZakopane
Period6/7/046/11/04

Fingerprint

Soft Sensor
Polymer Melts
Polymer melts
State Estimation
State estimation
Prediction
Sensors
Differential equations
Neural Networks
Spline Smoothing
Neural networks
Nonlinear Estimation
Algebraic Differential Equations
Polymerization
Knowledge-based
Model structures
Estimation Algorithms
Prior Knowledge
Splines
Nonlinear Differential Equations

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Feil, B., Abonyi, J., Pach, P., Nemeth, S., Arva, P., Nemeth, M., & Nagy, G. (2004). Semi-mechanistic models for state-estimation - Soft sensor for polymer melt index prediction. In L. Rutkowski, J. Siekmann, R. Tadeusiewicz, & L. A. Zadeh (Eds.), Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) (Vol. 3070, pp. 1111-1117)

Semi-mechanistic models for state-estimation - Soft sensor for polymer melt index prediction. / Feil, Balazs; Abonyi, J.; Pach, Peter; Nemeth, Sandor; Arva, Peter; Nemeth, Miklos; Nagy, Gabor.

Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). ed. / L. Rutkowski; J. Siekmann; R. Tadeusiewicz; L.A. Zadeh. Vol. 3070 2004. p. 1111-1117.

Research output: Conference contribution

Feil, B, Abonyi, J, Pach, P, Nemeth, S, Arva, P, Nemeth, M & Nagy, G 2004, Semi-mechanistic models for state-estimation - Soft sensor for polymer melt index prediction. in L Rutkowski, J Siekmann, R Tadeusiewicz & LA Zadeh (eds), Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). vol. 3070, pp. 1111-1117, 7th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2004, Zakopane, Poland, 6/7/04.
Feil B, Abonyi J, Pach P, Nemeth S, Arva P, Nemeth M et al. Semi-mechanistic models for state-estimation - Soft sensor for polymer melt index prediction. In Rutkowski L, Siekmann J, Tadeusiewicz R, Zadeh LA, editors, Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). Vol. 3070. 2004. p. 1111-1117
Feil, Balazs ; Abonyi, J. ; Pach, Peter ; Nemeth, Sandor ; Arva, Peter ; Nemeth, Miklos ; Nagy, Gabor. / Semi-mechanistic models for state-estimation - Soft sensor for polymer melt index prediction. Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). editor / L. Rutkowski ; J. Siekmann ; R. Tadeusiewicz ; L.A. Zadeh. Vol. 3070 2004. pp. 1111-1117
@inproceedings{2356f0e109bd4270ac1a26a64a40432e,
title = "Semi-mechanistic models for state-estimation - Soft sensor for polymer melt index prediction",
abstract = "Nonlinear state estimation is a useful approach to the monitoring of industrial (polymerization) processes. This paper investigates how this approach can be followed to the development of a soft sensor of the product quality (melt index). The bottleneck of the successful application of advanced state estimation algorithms is the identification of models that can accurately describe the process. This paper presents a semi-mechanistic modeling approach where neural networks describe the unknown phenomena of the system that cannot be formulated by prior knowledge based differential equations. Since in the presented semi-mechanistic model structure the neural network is a part of a nonlinear algebraic-differential equation set, there are no available direct input-output data to train the weights of the network. To handle this problem in this paper a simple, yet practically useful spline-smoothing based technique has been used. The results show that the developed semi-mechanistic model can be efficiently used for on-line state estimation.",
author = "Balazs Feil and J. Abonyi and Peter Pach and Sandor Nemeth and Peter Arva and Miklos Nemeth and Gabor Nagy",
year = "2004",
language = "English",
volume = "3070",
pages = "1111--1117",
editor = "L. Rutkowski and J. Siekmann and R. Tadeusiewicz and L.A. Zadeh",
booktitle = "Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)",

}

TY - GEN

T1 - Semi-mechanistic models for state-estimation - Soft sensor for polymer melt index prediction

AU - Feil, Balazs

AU - Abonyi, J.

AU - Pach, Peter

AU - Nemeth, Sandor

AU - Arva, Peter

AU - Nemeth, Miklos

AU - Nagy, Gabor

PY - 2004

Y1 - 2004

N2 - Nonlinear state estimation is a useful approach to the monitoring of industrial (polymerization) processes. This paper investigates how this approach can be followed to the development of a soft sensor of the product quality (melt index). The bottleneck of the successful application of advanced state estimation algorithms is the identification of models that can accurately describe the process. This paper presents a semi-mechanistic modeling approach where neural networks describe the unknown phenomena of the system that cannot be formulated by prior knowledge based differential equations. Since in the presented semi-mechanistic model structure the neural network is a part of a nonlinear algebraic-differential equation set, there are no available direct input-output data to train the weights of the network. To handle this problem in this paper a simple, yet practically useful spline-smoothing based technique has been used. The results show that the developed semi-mechanistic model can be efficiently used for on-line state estimation.

AB - Nonlinear state estimation is a useful approach to the monitoring of industrial (polymerization) processes. This paper investigates how this approach can be followed to the development of a soft sensor of the product quality (melt index). The bottleneck of the successful application of advanced state estimation algorithms is the identification of models that can accurately describe the process. This paper presents a semi-mechanistic modeling approach where neural networks describe the unknown phenomena of the system that cannot be formulated by prior knowledge based differential equations. Since in the presented semi-mechanistic model structure the neural network is a part of a nonlinear algebraic-differential equation set, there are no available direct input-output data to train the weights of the network. To handle this problem in this paper a simple, yet practically useful spline-smoothing based technique has been used. The results show that the developed semi-mechanistic model can be efficiently used for on-line state estimation.

UR - http://www.scopus.com/inward/record.url?scp=9444230588&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=9444230588&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:9444230588

VL - 3070

SP - 1111

EP - 1117

BT - Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)

A2 - Rutkowski, L.

A2 - Siekmann, J.

A2 - Tadeusiewicz, R.

A2 - Zadeh, L.A.

ER -