Publication detail

Adaptive Controllers by Using Neural Network Based Identification for Short Sampling Period

Pivoňka, P., Veleba, V., Ošmera, P.

Original Title

Adaptive Controllers by Using Neural Network Based Identification for Short Sampling Period

English Title

Adaptive Controllers by Using Neural Network Based Identification for Short Sampling Period

Type

conference paper

Language

en

Original Abstract

The use of short sampling period in adaptive control has not been described properly when controlling the real process by adaptive controller. The new approach to analysis of on-line identification methods based on one-step-ahead prediction clears up their sensitivity to disturbances in control loop. On one hand faster disturbance rejection due to short sampling period can be an advantage but on the other hand it brings us some practical problems. Particularly, quantization error and finite numerical precision of industrial controller must be considered in the real process control. We concentrate our attention on dealing with adverse effects that work on real-time identification of process, especially quantization. It is shown; that a neural network applied to on-line identification process produces more stable solution in the rapid sampling domain.

English abstract

The use of short sampling period in adaptive control has not been described properly when controlling the real process by adaptive controller. The new approach to analysis of on-line identification methods based on one-step-ahead prediction clears up their sensitivity to disturbances in control loop. On one hand faster disturbance rejection due to short sampling period can be an advantage but on the other hand it brings us some practical problems. Particularly, quantization error and finite numerical precision of industrial controller must be considered in the real process control. We concentrate our attention on dealing with adverse effects that work on real-time identification of process, especially quantization. It is shown; that a neural network applied to on-line identification process produces more stable solution in the rapid sampling domain.

Keywords

Adaptive Controllers, Neural Networks for Identification, Comparison of Identifications methods, Rapid Sampling Domain

RIV year

2006

Released

05.12.2006

Publisher

Nanyang Technological University

Location

Singapore

ISBN

1-4244-0342-1

Book

9th International Conference on Control, Automation, Robotics and Vision, IEEE ICARCV2006

Pages from

521

Pages to

526

Pages count

6

BibTex


@inproceedings{BUT22106,
  author="Petr {Pivoňka} and Václav {Veleba}",
  title="Adaptive Controllers by Using Neural Network Based Identification for Short Sampling Period",
  annote="The use of short sampling period in adaptive control has not been described properly when controlling the real process by adaptive controller. The new approach to analysis of on-line identification methods based on one-step-ahead prediction clears up their sensitivity to disturbances in control loop. On one hand faster disturbance rejection due to short sampling period can be an advantage but on the other hand it brings us some practical problems. Particularly, quantization error and finite numerical precision of industrial controller must be considered in the real process control. We concentrate our attention on dealing with adverse effects that work on real-time identification of process, especially quantization. It is shown; that a neural network applied to on-line identification process produces more stable solution in the rapid sampling domain.",
  address="Nanyang Technological University",
  booktitle="9th International Conference on Control, Automation, Robotics and Vision, IEEE ICARCV2006",
  chapter="22106",
  institution="Nanyang Technological University",
  year="2006",
  month="december",
  pages="521",
  publisher="Nanyang Technological University",
  type="conference paper"
}