Artificial-neural-network-based sensorless nonlinear control of induction motors

Miroslaw Wlas, Zbigniew Krzemiński, Jarosław Guziński, Haitham Abu-Rub, Hamid A. Toliyat

Research output: Contribution to journalArticle

90 Citations (Scopus)

Abstract

In this paper, two architectures of artificial neural networks (ANNs) are developed and used to correct the performance of sensorless nonlinear control of induction motor systems. Feed-forward multilayer perception, an Elman recurrent ANN, and a two-layer feedforward ANN is used in the control process. The method is based on the use of ANN to get an appropriate correction for improving the estimated speed. Simulation and experimental results were carried out for the proposed control system. An induction motor fed by voltage source inverter was used in the experimental system. A digital signal processor and field-programmable gate arrays were used to implement the control algorithm.

Original languageEnglish
Pages (from-to)520-528
Number of pages9
JournalIEEE Transactions on Energy Conversion
Volume20
Issue number3
DOIs
Publication statusPublished - Sep 2005
Externally publishedYes

Fingerprint

Induction motors
Neural networks
Digital signal processors
Field programmable gate arrays (FPGA)
Multilayers
Control systems
Electric potential

Keywords

  • Artificial neural networks (ANNs)
  • Digital signal processor (DSP)
  • Field-programmable gate arrays (FPGAs)
  • Induction motor
  • Nonlinear control
  • Observer system
  • Sensorless control

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Fuel Technology
  • Electrical and Electronic Engineering

Cite this

Artificial-neural-network-based sensorless nonlinear control of induction motors. / Wlas, Miroslaw; Krzemiński, Zbigniew; Guziński, Jarosław; Abu-Rub, Haitham; Toliyat, Hamid A.

In: IEEE Transactions on Energy Conversion, Vol. 20, No. 3, 09.2005, p. 520-528.

Research output: Contribution to journalArticle

Wlas, Miroslaw ; Krzemiński, Zbigniew ; Guziński, Jarosław ; Abu-Rub, Haitham ; Toliyat, Hamid A. / Artificial-neural-network-based sensorless nonlinear control of induction motors. In: IEEE Transactions on Energy Conversion. 2005 ; Vol. 20, No. 3. pp. 520-528.
@article{1451ceff82a640d4b0f3df2d6028bb43,
title = "Artificial-neural-network-based sensorless nonlinear control of induction motors",
abstract = "In this paper, two architectures of artificial neural networks (ANNs) are developed and used to correct the performance of sensorless nonlinear control of induction motor systems. Feed-forward multilayer perception, an Elman recurrent ANN, and a two-layer feedforward ANN is used in the control process. The method is based on the use of ANN to get an appropriate correction for improving the estimated speed. Simulation and experimental results were carried out for the proposed control system. An induction motor fed by voltage source inverter was used in the experimental system. A digital signal processor and field-programmable gate arrays were used to implement the control algorithm.",
keywords = "Artificial neural networks (ANNs), Digital signal processor (DSP), Field-programmable gate arrays (FPGAs), Induction motor, Nonlinear control, Observer system, Sensorless control",
author = "Miroslaw Wlas and Zbigniew Krzemiński and Jarosław Guziński and Haitham Abu-Rub and Toliyat, {Hamid A.}",
year = "2005",
month = "9",
doi = "10.1109/TEC.2005.847984",
language = "English",
volume = "20",
pages = "520--528",
journal = "IEEE Transactions on Energy Conversion",
issn = "0885-8969",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "3",

}

TY - JOUR

T1 - Artificial-neural-network-based sensorless nonlinear control of induction motors

AU - Wlas, Miroslaw

AU - Krzemiński, Zbigniew

AU - Guziński, Jarosław

AU - Abu-Rub, Haitham

AU - Toliyat, Hamid A.

PY - 2005/9

Y1 - 2005/9

N2 - In this paper, two architectures of artificial neural networks (ANNs) are developed and used to correct the performance of sensorless nonlinear control of induction motor systems. Feed-forward multilayer perception, an Elman recurrent ANN, and a two-layer feedforward ANN is used in the control process. The method is based on the use of ANN to get an appropriate correction for improving the estimated speed. Simulation and experimental results were carried out for the proposed control system. An induction motor fed by voltage source inverter was used in the experimental system. A digital signal processor and field-programmable gate arrays were used to implement the control algorithm.

AB - In this paper, two architectures of artificial neural networks (ANNs) are developed and used to correct the performance of sensorless nonlinear control of induction motor systems. Feed-forward multilayer perception, an Elman recurrent ANN, and a two-layer feedforward ANN is used in the control process. The method is based on the use of ANN to get an appropriate correction for improving the estimated speed. Simulation and experimental results were carried out for the proposed control system. An induction motor fed by voltage source inverter was used in the experimental system. A digital signal processor and field-programmable gate arrays were used to implement the control algorithm.

KW - Artificial neural networks (ANNs)

KW - Digital signal processor (DSP)

KW - Field-programmable gate arrays (FPGAs)

KW - Induction motor

KW - Nonlinear control

KW - Observer system

KW - Sensorless control

UR - http://www.scopus.com/inward/record.url?scp=24644498309&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=24644498309&partnerID=8YFLogxK

U2 - 10.1109/TEC.2005.847984

DO - 10.1109/TEC.2005.847984

M3 - Article

VL - 20

SP - 520

EP - 528

JO - IEEE Transactions on Energy Conversion

JF - IEEE Transactions on Energy Conversion

SN - 0885-8969

IS - 3

ER -