Machine Learning-Based Statistical Hypothesis Testing for Fault Detection

Radhia Fazai, Majdi Mansouri, Kamal Abodayeh, Mohamed Trabelsi, Hazem Nounou, Mohamed Nounou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper elaborates the development of machine learning approach merged with statistical hypothesis testing aimed at enhancing the operation of photovoltaic (PV) systems by developing intelligent PV fault detection framework. Fault detection in PV systems is important to ensure optimal energy harvesting and reliable power production because PV systems usually operate in a harsh outdoor environment and tend to suffer various faults. In this paper, therefore, special attention is paid to detection of various faults during different modes of operation. The proposed approach merges the benefits of machine learning technique (MLT) with statistical hypothesis testing to enhance the fault detection and monitoring of PV systems. The proposed technique will be effective in monitoring the PV faults under both normal and abnormal conditions. For the framework developed, the modeling phase is addressed using MLT and the faults are detected using the generalized likelihood ratio test (GLRT) chart. The MLT is used to compute the residuals monitored and the GLRT chart is applied to the monitored residuals evaluated for fault detection purposes. The developed MLT-based GLRT algorithm is implemented and validated using both simulated and real PV data. The results are evaluated in terms of false alarm rates (FAR), missed detection rates (MDR) and computation time.

Original languageEnglish
Title of host publication2019 4th Conference on Control and Fault Tolerant Systems, SysTol 2019
PublisherIEEE Computer Society
Pages38-43
Number of pages6
ISBN (Electronic)9781728103808
DOIs
Publication statusPublished - Sep 2019
Event4th Conference on Control and Fault Tolerant Systems, SysTol 2019 - Casablanca, Morocco
Duration: 18 Sep 201920 Sep 2019

Publication series

NameConference on Control and Fault-Tolerant Systems, SysTol
ISSN (Print)2162-1195
ISSN (Electronic)2162-1209

Conference

Conference4th Conference on Control and Fault Tolerant Systems, SysTol 2019
CountryMorocco
CityCasablanca
Period18/9/1920/9/19

Fingerprint

Fault detection
Learning systems
Testing
Monitoring
Energy harvesting

ASJC Scopus subject areas

  • Computer Science Applications
  • Hardware and Architecture
  • Software
  • Control and Systems Engineering

Cite this

Fazai, R., Mansouri, M., Abodayeh, K., Trabelsi, M., Nounou, H., & Nounou, M. (2019). Machine Learning-Based Statistical Hypothesis Testing for Fault Detection. In 2019 4th Conference on Control and Fault Tolerant Systems, SysTol 2019 (pp. 38-43). [8864776] (Conference on Control and Fault-Tolerant Systems, SysTol). IEEE Computer Society. https://doi.org/10.1109/SYSTOL.2019.8864776

Machine Learning-Based Statistical Hypothesis Testing for Fault Detection. / Fazai, Radhia; Mansouri, Majdi; Abodayeh, Kamal; Trabelsi, Mohamed; Nounou, Hazem; Nounou, Mohamed.

2019 4th Conference on Control and Fault Tolerant Systems, SysTol 2019. IEEE Computer Society, 2019. p. 38-43 8864776 (Conference on Control and Fault-Tolerant Systems, SysTol).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Fazai, R, Mansouri, M, Abodayeh, K, Trabelsi, M, Nounou, H & Nounou, M 2019, Machine Learning-Based Statistical Hypothesis Testing for Fault Detection. in 2019 4th Conference on Control and Fault Tolerant Systems, SysTol 2019., 8864776, Conference on Control and Fault-Tolerant Systems, SysTol, IEEE Computer Society, pp. 38-43, 4th Conference on Control and Fault Tolerant Systems, SysTol 2019, Casablanca, Morocco, 18/9/19. https://doi.org/10.1109/SYSTOL.2019.8864776
Fazai R, Mansouri M, Abodayeh K, Trabelsi M, Nounou H, Nounou M. Machine Learning-Based Statistical Hypothesis Testing for Fault Detection. In 2019 4th Conference on Control and Fault Tolerant Systems, SysTol 2019. IEEE Computer Society. 2019. p. 38-43. 8864776. (Conference on Control and Fault-Tolerant Systems, SysTol). https://doi.org/10.1109/SYSTOL.2019.8864776
Fazai, Radhia ; Mansouri, Majdi ; Abodayeh, Kamal ; Trabelsi, Mohamed ; Nounou, Hazem ; Nounou, Mohamed. / Machine Learning-Based Statistical Hypothesis Testing for Fault Detection. 2019 4th Conference on Control and Fault Tolerant Systems, SysTol 2019. IEEE Computer Society, 2019. pp. 38-43 (Conference on Control and Fault-Tolerant Systems, SysTol).
@inproceedings{9ffbcabb693f4f6688a64ea326e6a28d,
title = "Machine Learning-Based Statistical Hypothesis Testing for Fault Detection",
abstract = "This paper elaborates the development of machine learning approach merged with statistical hypothesis testing aimed at enhancing the operation of photovoltaic (PV) systems by developing intelligent PV fault detection framework. Fault detection in PV systems is important to ensure optimal energy harvesting and reliable power production because PV systems usually operate in a harsh outdoor environment and tend to suffer various faults. In this paper, therefore, special attention is paid to detection of various faults during different modes of operation. The proposed approach merges the benefits of machine learning technique (MLT) with statistical hypothesis testing to enhance the fault detection and monitoring of PV systems. The proposed technique will be effective in monitoring the PV faults under both normal and abnormal conditions. For the framework developed, the modeling phase is addressed using MLT and the faults are detected using the generalized likelihood ratio test (GLRT) chart. The MLT is used to compute the residuals monitored and the GLRT chart is applied to the monitored residuals evaluated for fault detection purposes. The developed MLT-based GLRT algorithm is implemented and validated using both simulated and real PV data. The results are evaluated in terms of false alarm rates (FAR), missed detection rates (MDR) and computation time.",
author = "Radhia Fazai and Majdi Mansouri and Kamal Abodayeh and Mohamed Trabelsi and Hazem Nounou and Mohamed Nounou",
year = "2019",
month = "9",
doi = "10.1109/SYSTOL.2019.8864776",
language = "English",
series = "Conference on Control and Fault-Tolerant Systems, SysTol",
publisher = "IEEE Computer Society",
pages = "38--43",
booktitle = "2019 4th Conference on Control and Fault Tolerant Systems, SysTol 2019",

}

TY - GEN

T1 - Machine Learning-Based Statistical Hypothesis Testing for Fault Detection

AU - Fazai, Radhia

AU - Mansouri, Majdi

AU - Abodayeh, Kamal

AU - Trabelsi, Mohamed

AU - Nounou, Hazem

AU - Nounou, Mohamed

PY - 2019/9

Y1 - 2019/9

N2 - This paper elaborates the development of machine learning approach merged with statistical hypothesis testing aimed at enhancing the operation of photovoltaic (PV) systems by developing intelligent PV fault detection framework. Fault detection in PV systems is important to ensure optimal energy harvesting and reliable power production because PV systems usually operate in a harsh outdoor environment and tend to suffer various faults. In this paper, therefore, special attention is paid to detection of various faults during different modes of operation. The proposed approach merges the benefits of machine learning technique (MLT) with statistical hypothesis testing to enhance the fault detection and monitoring of PV systems. The proposed technique will be effective in monitoring the PV faults under both normal and abnormal conditions. For the framework developed, the modeling phase is addressed using MLT and the faults are detected using the generalized likelihood ratio test (GLRT) chart. The MLT is used to compute the residuals monitored and the GLRT chart is applied to the monitored residuals evaluated for fault detection purposes. The developed MLT-based GLRT algorithm is implemented and validated using both simulated and real PV data. The results are evaluated in terms of false alarm rates (FAR), missed detection rates (MDR) and computation time.

AB - This paper elaborates the development of machine learning approach merged with statistical hypothesis testing aimed at enhancing the operation of photovoltaic (PV) systems by developing intelligent PV fault detection framework. Fault detection in PV systems is important to ensure optimal energy harvesting and reliable power production because PV systems usually operate in a harsh outdoor environment and tend to suffer various faults. In this paper, therefore, special attention is paid to detection of various faults during different modes of operation. The proposed approach merges the benefits of machine learning technique (MLT) with statistical hypothesis testing to enhance the fault detection and monitoring of PV systems. The proposed technique will be effective in monitoring the PV faults under both normal and abnormal conditions. For the framework developed, the modeling phase is addressed using MLT and the faults are detected using the generalized likelihood ratio test (GLRT) chart. The MLT is used to compute the residuals monitored and the GLRT chart is applied to the monitored residuals evaluated for fault detection purposes. The developed MLT-based GLRT algorithm is implemented and validated using both simulated and real PV data. The results are evaluated in terms of false alarm rates (FAR), missed detection rates (MDR) and computation time.

UR - http://www.scopus.com/inward/record.url?scp=85077613579&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85077613579&partnerID=8YFLogxK

U2 - 10.1109/SYSTOL.2019.8864776

DO - 10.1109/SYSTOL.2019.8864776

M3 - Conference contribution

AN - SCOPUS:85077613579

T3 - Conference on Control and Fault-Tolerant Systems, SysTol

SP - 38

EP - 43

BT - 2019 4th Conference on Control and Fault Tolerant Systems, SysTol 2019

PB - IEEE Computer Society

ER -