Fault Diagnosis of Inverter Current Sensor Using Artificial Neural Network Considering Out-of-distribution

Jae Hoon Shim; Kahyun Lee; Jung Ik Ha

doi:10.1109/ECCE-Asia49820.2021.9479085

Fault Diagnosis of Inverter Current Sensor Using Artificial Neural Network Considering Out-of-distribution

Jae Hoon Shim, Kahyun Lee, Jung Ik Ha

Electronic and Electrical Engineering

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

7 Scopus citations

Abstract

This paper presents an Artificial Neural Network (ANN)-based approach for effectively diagnosing current sensor faults in inverters under various operating conditions. In this approach, additional techniques to make a robust fault classifier for untrained out-of-distribution (OOD) data are considered. The ANN is trained and evaluated through the Python TensorFlow library. The fault diagnosis accuracy is measured using the confusion matrix on the in-distribution dataset (current sensor faults data), and the degree of performance improvement on the OOD dataset is evaluated through the Area Under the Receiver Operating Characteristic (AUROC). The proposed method has been verified to have a classification accuracy of 98.94% on the in-distribution dataset and improve the AUROC from 0.9170 to 0.9866.

Original language	English
Title of host publication	Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1520-1525
Number of pages	6
ISBN (Electronic)	9781728163444
DOIs	https://doi.org/10.1109/ECCE-Asia49820.2021.9479085
State	Published - 24 May 2021
Event	12th IEEE Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021 - Virtual, Singapore, Singapore Duration: 24 May 2021 → 27 May 2021

Publication series

Name	Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021

Conference

Conference	12th IEEE Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021
Country/Territory	Singapore
City	Virtual, Singapore
Period	24/05/21 → 27/05/21

Bibliographical note

Publisher Copyright:
© 2021 IEEE.

Keywords

artificial neural network
fault detection
fault diagnosis
in-distribution data
out-of distribution data

Access to Document

10.1109/ECCE-Asia49820.2021.9479085

Cite this

Shim, J. H., Lee, K., & Ha, J. I. (2021). Fault Diagnosis of Inverter Current Sensor Using Artificial Neural Network Considering Out-of-distribution. In Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021 (pp. 1520-1525). Article 9479085 (Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ECCE-Asia49820.2021.9479085

Shim, Jae Hoon ; Lee, Kahyun ; Ha, Jung Ik. / Fault Diagnosis of Inverter Current Sensor Using Artificial Neural Network Considering Out-of-distribution. Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 1520-1525 (Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021).

@inproceedings{1a66296a464d4b2fa8bf3e0c0675f02c,

title = "Fault Diagnosis of Inverter Current Sensor Using Artificial Neural Network Considering Out-of-distribution",

abstract = "This paper presents an Artificial Neural Network (ANN)-based approach for effectively diagnosing current sensor faults in inverters under various operating conditions. In this approach, additional techniques to make a robust fault classifier for untrained out-of-distribution (OOD) data are considered. The ANN is trained and evaluated through the Python TensorFlow library. The fault diagnosis accuracy is measured using the confusion matrix on the in-distribution dataset (current sensor faults data), and the degree of performance improvement on the OOD dataset is evaluated through the Area Under the Receiver Operating Characteristic (AUROC). The proposed method has been verified to have a classification accuracy of 98.94% on the in-distribution dataset and improve the AUROC from 0.9170 to 0.9866.",

keywords = "artificial neural network, fault detection, fault diagnosis, in-distribution data, out-of distribution data",

author = "Shim, {Jae Hoon} and Kahyun Lee and Ha, {Jung Ik}",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 12th IEEE Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021 ; Conference date: 24-05-2021 Through 27-05-2021",

year = "2021",

month = may,

day = "24",

doi = "10.1109/ECCE-Asia49820.2021.9479085",

language = "English",

series = "Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1520--1525",

booktitle = "Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021",

}

Shim, JH, Lee, K & Ha, JI 2021, Fault Diagnosis of Inverter Current Sensor Using Artificial Neural Network Considering Out-of-distribution. in Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021., 9479085, Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021, Institute of Electrical and Electronics Engineers Inc., pp. 1520-1525, 12th IEEE Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021, Virtual, Singapore, Singapore, 24/05/21. https://doi.org/10.1109/ECCE-Asia49820.2021.9479085

Fault Diagnosis of Inverter Current Sensor Using Artificial Neural Network Considering Out-of-distribution. / Shim, Jae Hoon; Lee, Kahyun; Ha, Jung Ik.
Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 1520-1525 9479085 (Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021).

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

TY - GEN

T1 - Fault Diagnosis of Inverter Current Sensor Using Artificial Neural Network Considering Out-of-distribution

AU - Shim, Jae Hoon

AU - Lee, Kahyun

AU - Ha, Jung Ik

PY - 2021/5/24

Y1 - 2021/5/24

N2 - This paper presents an Artificial Neural Network (ANN)-based approach for effectively diagnosing current sensor faults in inverters under various operating conditions. In this approach, additional techniques to make a robust fault classifier for untrained out-of-distribution (OOD) data are considered. The ANN is trained and evaluated through the Python TensorFlow library. The fault diagnosis accuracy is measured using the confusion matrix on the in-distribution dataset (current sensor faults data), and the degree of performance improvement on the OOD dataset is evaluated through the Area Under the Receiver Operating Characteristic (AUROC). The proposed method has been verified to have a classification accuracy of 98.94% on the in-distribution dataset and improve the AUROC from 0.9170 to 0.9866.

AB - This paper presents an Artificial Neural Network (ANN)-based approach for effectively diagnosing current sensor faults in inverters under various operating conditions. In this approach, additional techniques to make a robust fault classifier for untrained out-of-distribution (OOD) data are considered. The ANN is trained and evaluated through the Python TensorFlow library. The fault diagnosis accuracy is measured using the confusion matrix on the in-distribution dataset (current sensor faults data), and the degree of performance improvement on the OOD dataset is evaluated through the Area Under the Receiver Operating Characteristic (AUROC). The proposed method has been verified to have a classification accuracy of 98.94% on the in-distribution dataset and improve the AUROC from 0.9170 to 0.9866.

KW - artificial neural network

KW - fault detection

KW - fault diagnosis

KW - in-distribution data

KW - out-of distribution data

UR - http://www.scopus.com/inward/record.url?scp=85114196707&partnerID=8YFLogxK

U2 - 10.1109/ECCE-Asia49820.2021.9479085

DO - 10.1109/ECCE-Asia49820.2021.9479085

M3 - Conference contribution

AN - SCOPUS:85114196707

T3 - Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021

SP - 1520

EP - 1525

BT - Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 12th IEEE Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021

Y2 - 24 May 2021 through 27 May 2021

ER -

Shim JH, Lee K, Ha JI. Fault Diagnosis of Inverter Current Sensor Using Artificial Neural Network Considering Out-of-distribution. In Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 1520-1525. 9479085. (Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021). doi: 10.1109/ECCE-Asia49820.2021.9479085

Fault Diagnosis of Inverter Current Sensor Using Artificial Neural Network Considering Out-of-distribution

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this