Neural network based diagnosis system for looper height controller of hot strip mills

Yoshihiro Abe, Masami Konishi, Jun Imai

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

In this study, auto tuning of PID control gains in hot strip looper controller is made based on a recurrent neural network (RNN) model. Further, a neural network emulator (Neuro emulator) is employed to model the characteristics of looper dynamics. Combining Neuro emulator and the RNN model, an auto tuning system of PID control gains is constructed. As the inputs to RNN, plural evaluation functions which reflect individual preference of human experts. Self learning mechanism is embedded in the RNN model which enables adaptation both of rolling characteristics. However, over time, deterioration of mechanical characteristics and the control system will be induced. To overcome the problem, it is necessary to diagnose the true cause of failure and to compensate for it. For this purpose, the hierarchical neural network (HNN) is built into the auto tuning of PID control gains. The HNN model, which enables compensation to the deterioration of mill system, can estimate current system parameters such as control gains and mill constants. In developing the advanced control system combining HNN, RNN and Neuro Emulator, we aim for the realization of the simultaneous operations of tuning, diagnosis and estimation of the plant like skillful human experts. Through numerical experiments, the effect of the proposed method is ascertained.

Original languageEnglish
Pages (from-to)919-935
Number of pages17
JournalInternational Journal of Innovative Computing, Information and Control
Volume3
Issue number4
Publication statusPublished - Aug 2007

Fingerprint

Strip mills
Recurrent neural networks
Recurrent Neural Networks
Strip
Auto-tuning
Neural Network Model
Hierarchical Networks
Neural Networks
PID Control
Three term control systems
Neural networks
Controller
Tuning
Controllers
Deterioration
Control System
Control systems
Self-learning
Function evaluation
Gain control

Keywords

  • Hierarchical neural network (HNN)
  • Hot strip rolling
  • Looper height control
  • Neural network emulator (Neuro emulator)
  • PID control
  • Recurrent neural network (RNN)

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Information Systems
  • Software
  • Theoretical Computer Science

Cite this

Neural network based diagnosis system for looper height controller of hot strip mills. / Abe, Yoshihiro; Konishi, Masami; Imai, Jun.

In: International Journal of Innovative Computing, Information and Control, Vol. 3, No. 4, 08.2007, p. 919-935.

Research output: Contribution to journalArticle

@article{f0e39bf6218c45c192d07a5802aa639b,
title = "Neural network based diagnosis system for looper height controller of hot strip mills",
abstract = "In this study, auto tuning of PID control gains in hot strip looper controller is made based on a recurrent neural network (RNN) model. Further, a neural network emulator (Neuro emulator) is employed to model the characteristics of looper dynamics. Combining Neuro emulator and the RNN model, an auto tuning system of PID control gains is constructed. As the inputs to RNN, plural evaluation functions which reflect individual preference of human experts. Self learning mechanism is embedded in the RNN model which enables adaptation both of rolling characteristics. However, over time, deterioration of mechanical characteristics and the control system will be induced. To overcome the problem, it is necessary to diagnose the true cause of failure and to compensate for it. For this purpose, the hierarchical neural network (HNN) is built into the auto tuning of PID control gains. The HNN model, which enables compensation to the deterioration of mill system, can estimate current system parameters such as control gains and mill constants. In developing the advanced control system combining HNN, RNN and Neuro Emulator, we aim for the realization of the simultaneous operations of tuning, diagnosis and estimation of the plant like skillful human experts. Through numerical experiments, the effect of the proposed method is ascertained.",
keywords = "Hierarchical neural network (HNN), Hot strip rolling, Looper height control, Neural network emulator (Neuro emulator), PID control, Recurrent neural network (RNN)",
author = "Yoshihiro Abe and Masami Konishi and Jun Imai",
year = "2007",
month = "8",
language = "English",
volume = "3",
pages = "919--935",
journal = "International Journal of Innovative Computing, Information and Control",
issn = "1349-4198",
publisher = "IJICIC Editorial Office",
number = "4",

}

TY - JOUR

T1 - Neural network based diagnosis system for looper height controller of hot strip mills

AU - Abe, Yoshihiro

AU - Konishi, Masami

AU - Imai, Jun

PY - 2007/8

Y1 - 2007/8

N2 - In this study, auto tuning of PID control gains in hot strip looper controller is made based on a recurrent neural network (RNN) model. Further, a neural network emulator (Neuro emulator) is employed to model the characteristics of looper dynamics. Combining Neuro emulator and the RNN model, an auto tuning system of PID control gains is constructed. As the inputs to RNN, plural evaluation functions which reflect individual preference of human experts. Self learning mechanism is embedded in the RNN model which enables adaptation both of rolling characteristics. However, over time, deterioration of mechanical characteristics and the control system will be induced. To overcome the problem, it is necessary to diagnose the true cause of failure and to compensate for it. For this purpose, the hierarchical neural network (HNN) is built into the auto tuning of PID control gains. The HNN model, which enables compensation to the deterioration of mill system, can estimate current system parameters such as control gains and mill constants. In developing the advanced control system combining HNN, RNN and Neuro Emulator, we aim for the realization of the simultaneous operations of tuning, diagnosis and estimation of the plant like skillful human experts. Through numerical experiments, the effect of the proposed method is ascertained.

AB - In this study, auto tuning of PID control gains in hot strip looper controller is made based on a recurrent neural network (RNN) model. Further, a neural network emulator (Neuro emulator) is employed to model the characteristics of looper dynamics. Combining Neuro emulator and the RNN model, an auto tuning system of PID control gains is constructed. As the inputs to RNN, plural evaluation functions which reflect individual preference of human experts. Self learning mechanism is embedded in the RNN model which enables adaptation both of rolling characteristics. However, over time, deterioration of mechanical characteristics and the control system will be induced. To overcome the problem, it is necessary to diagnose the true cause of failure and to compensate for it. For this purpose, the hierarchical neural network (HNN) is built into the auto tuning of PID control gains. The HNN model, which enables compensation to the deterioration of mill system, can estimate current system parameters such as control gains and mill constants. In developing the advanced control system combining HNN, RNN and Neuro Emulator, we aim for the realization of the simultaneous operations of tuning, diagnosis and estimation of the plant like skillful human experts. Through numerical experiments, the effect of the proposed method is ascertained.

KW - Hierarchical neural network (HNN)

KW - Hot strip rolling

KW - Looper height control

KW - Neural network emulator (Neuro emulator)

KW - PID control

KW - Recurrent neural network (RNN)

UR - http://www.scopus.com/inward/record.url?scp=63549107108&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=63549107108&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:63549107108

VL - 3

SP - 919

EP - 935

JO - International Journal of Innovative Computing, Information and Control

JF - International Journal of Innovative Computing, Information and Control

SN - 1349-4198

IS - 4

ER -