Online reduced kernel GLRT technique for improved fault detection in photovoltaic systems

R. Fezai, Majdi Mansouri, M. Trabelsi, M. Hajji, H. Nounou, M. Nounou

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper proposes an effective kernel generalized likelihood ratio test (KGLRT) technique for fault detection in Photovoltaic (PV) systems. The proposed technique is considered as an improvement of the conventional KGLRT with extended online capabilities and lower computational complexity. The proposed online reduced KGLRT (OR-KGLRT) is based on transforming the process data into a higher dimensional space (where the data becomes linear), which makes the kernel-based scheme attractive for modeling nonlinear systems. The performance of the proposed method is evaluated and compared to the conventional KGLRT statistic using a simulated PV data. Both techniques are applied to detect single and multiple failures (including Bypass, Mismatch, Mix and Shading failures). The selected performance criteria are the good detection rate (GDR), false alarm rate (FAR), and computation time (CT). Simulation results show superior detection efficiency of the proposed approach compared to the conventional KGLRT statistic in terms of GDR, FAR and CT.

Original languageEnglish
Pages (from-to)1133-1154
Number of pages22
JournalEnergy
Volume179
DOIs
Publication statusPublished - 15 Jul 2019

Fingerprint

Fault detection
Statistics
Nonlinear systems
Computational complexity

Keywords

  • Fault detection
  • Kernel generalized likelihood ratio test (KGLRT)
  • Kernel principal component analysis (KPCA)
  • Online reduced GLRT (OR-GLRT)
  • Photovoltaic (PV) system

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Building and Construction
  • Pollution
  • Mechanical Engineering
  • Industrial and Manufacturing Engineering
  • Electrical and Electronic Engineering

Cite this

Online reduced kernel GLRT technique for improved fault detection in photovoltaic systems. / Fezai, R.; Mansouri, Majdi; Trabelsi, M.; Hajji, M.; Nounou, H.; Nounou, M.

In: Energy, Vol. 179, 15.07.2019, p. 1133-1154.

Research output: Contribution to journalArticle

Fezai, R. ; Mansouri, Majdi ; Trabelsi, M. ; Hajji, M. ; Nounou, H. ; Nounou, M. / Online reduced kernel GLRT technique for improved fault detection in photovoltaic systems. In: Energy. 2019 ; Vol. 179. pp. 1133-1154.
@article{70a3aaa9a3fe4fd5b8b00448a0a1f188,
title = "Online reduced kernel GLRT technique for improved fault detection in photovoltaic systems",
abstract = "This paper proposes an effective kernel generalized likelihood ratio test (KGLRT) technique for fault detection in Photovoltaic (PV) systems. The proposed technique is considered as an improvement of the conventional KGLRT with extended online capabilities and lower computational complexity. The proposed online reduced KGLRT (OR-KGLRT) is based on transforming the process data into a higher dimensional space (where the data becomes linear), which makes the kernel-based scheme attractive for modeling nonlinear systems. The performance of the proposed method is evaluated and compared to the conventional KGLRT statistic using a simulated PV data. Both techniques are applied to detect single and multiple failures (including Bypass, Mismatch, Mix and Shading failures). The selected performance criteria are the good detection rate (GDR), false alarm rate (FAR), and computation time (CT). Simulation results show superior detection efficiency of the proposed approach compared to the conventional KGLRT statistic in terms of GDR, FAR and CT.",
keywords = "Fault detection, Kernel generalized likelihood ratio test (KGLRT), Kernel principal component analysis (KPCA), Online reduced GLRT (OR-GLRT), Photovoltaic (PV) system",
author = "R. Fezai and Majdi Mansouri and M. Trabelsi and M. Hajji and H. Nounou and M. Nounou",
year = "2019",
month = "7",
day = "15",
doi = "10.1016/j.energy.2019.05.029",
language = "English",
volume = "179",
pages = "1133--1154",
journal = "Energy",
issn = "0360-5442",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Online reduced kernel GLRT technique for improved fault detection in photovoltaic systems

AU - Fezai, R.

AU - Mansouri, Majdi

AU - Trabelsi, M.

AU - Hajji, M.

AU - Nounou, H.

AU - Nounou, M.

PY - 2019/7/15

Y1 - 2019/7/15

N2 - This paper proposes an effective kernel generalized likelihood ratio test (KGLRT) technique for fault detection in Photovoltaic (PV) systems. The proposed technique is considered as an improvement of the conventional KGLRT with extended online capabilities and lower computational complexity. The proposed online reduced KGLRT (OR-KGLRT) is based on transforming the process data into a higher dimensional space (where the data becomes linear), which makes the kernel-based scheme attractive for modeling nonlinear systems. The performance of the proposed method is evaluated and compared to the conventional KGLRT statistic using a simulated PV data. Both techniques are applied to detect single and multiple failures (including Bypass, Mismatch, Mix and Shading failures). The selected performance criteria are the good detection rate (GDR), false alarm rate (FAR), and computation time (CT). Simulation results show superior detection efficiency of the proposed approach compared to the conventional KGLRT statistic in terms of GDR, FAR and CT.

AB - This paper proposes an effective kernel generalized likelihood ratio test (KGLRT) technique for fault detection in Photovoltaic (PV) systems. The proposed technique is considered as an improvement of the conventional KGLRT with extended online capabilities and lower computational complexity. The proposed online reduced KGLRT (OR-KGLRT) is based on transforming the process data into a higher dimensional space (where the data becomes linear), which makes the kernel-based scheme attractive for modeling nonlinear systems. The performance of the proposed method is evaluated and compared to the conventional KGLRT statistic using a simulated PV data. Both techniques are applied to detect single and multiple failures (including Bypass, Mismatch, Mix and Shading failures). The selected performance criteria are the good detection rate (GDR), false alarm rate (FAR), and computation time (CT). Simulation results show superior detection efficiency of the proposed approach compared to the conventional KGLRT statistic in terms of GDR, FAR and CT.

KW - Fault detection

KW - Kernel generalized likelihood ratio test (KGLRT)

KW - Kernel principal component analysis (KPCA)

KW - Online reduced GLRT (OR-GLRT)

KW - Photovoltaic (PV) system

UR - http://www.scopus.com/inward/record.url?scp=85065698741&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065698741&partnerID=8YFLogxK

U2 - 10.1016/j.energy.2019.05.029

DO - 10.1016/j.energy.2019.05.029

M3 - Article

AN - SCOPUS:85065698741

VL - 179

SP - 1133

EP - 1154

JO - Energy

JF - Energy

SN - 0360-5442

ER -