Active Power Filter Design Using Neural Network with a Variable Step-Size

Alejandro Alvarez; Joseph Cherre; Christian Flores; Julien Noel

doi:10.1007/978-3-031-04435-9_12

Active Power Filter Design Using Neural Network with a Variable Step-Size

Alejandro Alvarez, Joseph Cherre, Christian Flores, Julien Noel

Department of Electrical and Mechatronics Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Harmonic elimination in three-phase power systems due to non-linear loads in electrical grids is an essential topic in power quality nowadays. The rise of electronics and electrical devices increases the sensitivity of the network. The current research approaches this topic by proposing the design and simulation of a neural network with a modified LMS (Least Mean Square) algorithm, including variable step-size in the cost function to achieve faster convergence and overcoming problems in drifting parameters. We use a three-phase active power filter (APF) that relies on a current compensation control for testing purposes. The APF consists of a hysteresis control block and a current-controlled voltage source inverter computed by our enhanced ADALINE (Adaptative Linear Element) neural network. Finally, a comparison is made with another algorithm based on non-variable step size. Results show a 3 % reduction in THD (Total Harmonic Distortion) parameter and a 0.2 s faster response time, concluding the improvement in performance parameters of the power system.

Original language	English
Title of host publication	Proceedings of the 7th Brazilian Technology Symposium, BTSym 2021 - Emerging Trends in Human Smart and Sustainable Future of Cities Volume 1
Editors	Yuzo Iano, Osamu Saotome, Guillermo Leopoldo Kemper Vásquez, Claudia Cotrim Pezzuto, Rangel Arthur, Gabriel Gomes de Oliveira
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	123-132
Number of pages	10
ISBN (Print)	9783031044342
DOIs	https://doi.org/10.1007/978-3-031-04435-9_12
State	Published - 2023
Event	7th Brazilian Technology Symposium, BTSym 2021 - Virtual, Online Duration: 8 Nov 2021 → 10 Nov 2021

Publication series

Name	Smart Innovation, Systems and Technologies
Volume	207 SIST
ISSN (Print)	2190-3018
ISSN (Electronic)	2190-3026

Conference

Conference	7th Brazilian Technology Symposium, BTSym 2021
City	Virtual, Online
Period	8/11/21 → 10/11/21

Keywords

ADALINE
APF
Harmonic distortion
Hysteresis
MATLAB
Simulink
THD
Three-phase power system

Access to Document

10.1007/978-3-031-04435-9_12

Cite this

Alvarez, A., Cherre, J., Flores, C., & Noel, J. (2023). Active Power Filter Design Using Neural Network with a Variable Step-Size. In Y. Iano, O. Saotome, G. L. Kemper Vásquez, C. Cotrim Pezzuto, R. Arthur, & G. Gomes de Oliveira (Eds.), Proceedings of the 7th Brazilian Technology Symposium, BTSym 2021 - Emerging Trends in Human Smart and Sustainable Future of Cities Volume 1 (pp. 123-132). (Smart Innovation, Systems and Technologies; Vol. 207 SIST). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-04435-9_12

Alvarez, Alejandro ; Cherre, Joseph ; Flores, Christian et al. / Active Power Filter Design Using Neural Network with a Variable Step-Size. Proceedings of the 7th Brazilian Technology Symposium, BTSym 2021 - Emerging Trends in Human Smart and Sustainable Future of Cities Volume 1. editor / Yuzo Iano ; Osamu Saotome ; Guillermo Leopoldo Kemper Vásquez ; Claudia Cotrim Pezzuto ; Rangel Arthur ; Gabriel Gomes de Oliveira. Springer Science and Business Media Deutschland GmbH, 2023. pp. 123-132 (Smart Innovation, Systems and Technologies).

@inproceedings{fb7d95ad68a243bc86714133b5705cc7,

title = "Active Power Filter Design Using Neural Network with a Variable Step-Size",

abstract = "Harmonic elimination in three-phase power systems due to non-linear loads in electrical grids is an essential topic in power quality nowadays. The rise of electronics and electrical devices increases the sensitivity of the network. The current research approaches this topic by proposing the design and simulation of a neural network with a modified LMS (Least Mean Square) algorithm, including variable step-size in the cost function to achieve faster convergence and overcoming problems in drifting parameters. We use a three-phase active power filter (APF) that relies on a current compensation control for testing purposes. The APF consists of a hysteresis control block and a current-controlled voltage source inverter computed by our enhanced ADALINE (Adaptative Linear Element) neural network. Finally, a comparison is made with another algorithm based on non-variable step size. Results show a 3 % reduction in THD (Total Harmonic Distortion) parameter and a 0.2 s faster response time, concluding the improvement in performance parameters of the power system.",

keywords = "ADALINE, APF, Harmonic distortion, Hysteresis, MATLAB, Simulink, THD, Three-phase power system",

author = "Alejandro Alvarez and Joseph Cherre and Christian Flores and Julien Noel",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 7th Brazilian Technology Symposium, BTSym 2021 ; Conference date: 08-11-2021 Through 10-11-2021",

year = "2023",

doi = "10.1007/978-3-031-04435-9_12",

language = "English",

isbn = "9783031044342",

series = "Smart Innovation, Systems and Technologies",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "123--132",

editor = "Yuzo Iano and Osamu Saotome and {Kemper V{\'a}squez}, {Guillermo Leopoldo} and {Cotrim Pezzuto}, Claudia and Rangel Arthur and {Gomes de Oliveira}, Gabriel",

booktitle = "Proceedings of the 7th Brazilian Technology Symposium, BTSym 2021 - Emerging Trends in Human Smart and Sustainable Future of Cities Volume 1",

}

Alvarez, A, Cherre, J, Flores, C & Noel, J 2023, Active Power Filter Design Using Neural Network with a Variable Step-Size. in Y Iano, O Saotome, GL Kemper Vásquez, C Cotrim Pezzuto, R Arthur & G Gomes de Oliveira (eds), Proceedings of the 7th Brazilian Technology Symposium, BTSym 2021 - Emerging Trends in Human Smart and Sustainable Future of Cities Volume 1. Smart Innovation, Systems and Technologies, vol. 207 SIST, Springer Science and Business Media Deutschland GmbH, pp. 123-132, 7th Brazilian Technology Symposium, BTSym 2021, Virtual, Online, 8/11/21. https://doi.org/10.1007/978-3-031-04435-9_12

Active Power Filter Design Using Neural Network with a Variable Step-Size. / Alvarez, Alejandro; Cherre, Joseph; Flores, Christian et al.
Proceedings of the 7th Brazilian Technology Symposium, BTSym 2021 - Emerging Trends in Human Smart and Sustainable Future of Cities Volume 1. ed. / Yuzo Iano; Osamu Saotome; Guillermo Leopoldo Kemper Vásquez; Claudia Cotrim Pezzuto; Rangel Arthur; Gabriel Gomes de Oliveira. Springer Science and Business Media Deutschland GmbH, 2023. p. 123-132 (Smart Innovation, Systems and Technologies; Vol. 207 SIST).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Active Power Filter Design Using Neural Network with a Variable Step-Size

AU - Alvarez, Alejandro

AU - Cherre, Joseph

AU - Flores, Christian

AU - Noel, Julien

PY - 2023

Y1 - 2023

N2 - Harmonic elimination in three-phase power systems due to non-linear loads in electrical grids is an essential topic in power quality nowadays. The rise of electronics and electrical devices increases the sensitivity of the network. The current research approaches this topic by proposing the design and simulation of a neural network with a modified LMS (Least Mean Square) algorithm, including variable step-size in the cost function to achieve faster convergence and overcoming problems in drifting parameters. We use a three-phase active power filter (APF) that relies on a current compensation control for testing purposes. The APF consists of a hysteresis control block and a current-controlled voltage source inverter computed by our enhanced ADALINE (Adaptative Linear Element) neural network. Finally, a comparison is made with another algorithm based on non-variable step size. Results show a 3 % reduction in THD (Total Harmonic Distortion) parameter and a 0.2 s faster response time, concluding the improvement in performance parameters of the power system.

AB - Harmonic elimination in three-phase power systems due to non-linear loads in electrical grids is an essential topic in power quality nowadays. The rise of electronics and electrical devices increases the sensitivity of the network. The current research approaches this topic by proposing the design and simulation of a neural network with a modified LMS (Least Mean Square) algorithm, including variable step-size in the cost function to achieve faster convergence and overcoming problems in drifting parameters. We use a three-phase active power filter (APF) that relies on a current compensation control for testing purposes. The APF consists of a hysteresis control block and a current-controlled voltage source inverter computed by our enhanced ADALINE (Adaptative Linear Element) neural network. Finally, a comparison is made with another algorithm based on non-variable step size. Results show a 3 % reduction in THD (Total Harmonic Distortion) parameter and a 0.2 s faster response time, concluding the improvement in performance parameters of the power system.

KW - ADALINE

KW - APF

KW - Harmonic distortion

KW - Hysteresis

KW - MATLAB

KW - Simulink

KW - THD

KW - Three-phase power system

UR - http://www.scopus.com/inward/record.url?scp=85135072773&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-04435-9_12

DO - 10.1007/978-3-031-04435-9_12

M3 - Conference contribution

AN - SCOPUS:85135072773

SN - 9783031044342

T3 - Smart Innovation, Systems and Technologies

SP - 123

EP - 132

BT - Proceedings of the 7th Brazilian Technology Symposium, BTSym 2021 - Emerging Trends in Human Smart and Sustainable Future of Cities Volume 1

A2 - Iano, Yuzo

A2 - Saotome, Osamu

A2 - Kemper Vásquez, Guillermo Leopoldo

A2 - Cotrim Pezzuto, Claudia

A2 - Arthur, Rangel

A2 - Gomes de Oliveira, Gabriel

PB - Springer Science and Business Media Deutschland GmbH

T2 - 7th Brazilian Technology Symposium, BTSym 2021

Y2 - 8 November 2021 through 10 November 2021

ER -

Alvarez A, Cherre J, Flores C , Noel J. Active Power Filter Design Using Neural Network with a Variable Step-Size. In Iano Y, Saotome O, Kemper Vásquez GL, Cotrim Pezzuto C, Arthur R, Gomes de Oliveira G, editors, Proceedings of the 7th Brazilian Technology Symposium, BTSym 2021 - Emerging Trends in Human Smart and Sustainable Future of Cities Volume 1. Springer Science and Business Media Deutschland GmbH. 2023. p. 123-132. (Smart Innovation, Systems and Technologies). doi: 10.1007/978-3-031-04435-9_12