Employing the Generative Adversarial Networks (GAN) for Reliability Assessment of Converters

Amirali Davoodi; Saeed Peyghami; Yongheng Yang; Tomislav Dragicevic; Frede Blaabjerg

doi:10.1109/ECCE47101.2021.9595614

Employing the Generative Adversarial Networks (GAN) for Reliability Assessment of Converters

Amirali Davoodi
, Saeed Peyghami
, Yongheng Yang
, Tomislav Dragicevic
, Frede Blaabjerg

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

Abstract

Mission profiles are widely used for the reliability analysis of power converters. Typically, to assess the converter reliability, long-term (e.g., one year) mission profiles are adopted, and it is assumed that the profiles will be repeated in future years. However, due to mission profile uncertainties, the assumption can introduce considerable errors in the estimated reliability. In this paper, the errors introduced by the above assumption are studied in detail. Furthermore, to tackle this challenge, the paper proposes using the Generative Adversarial Networks (GAN) to generate unique mission profile scenarios that capture the temporal and probabilistic properties of the real profiles. In this regard, the effectiveness of using the GAN-generated profiles to improve the accuracy of the estimated reliability is demonstrated.

Original language	English
Title of host publication	Proceedings of 2021 IEEE Energy Conversion Congress and Exposition
Publisher	IEEE
Publication date	2021
Pages	3623-3629
ISBN (Print)	978-1-7281-6128-0
DOIs	https://doi.org/10.1109/ECCE47101.2021.9595614
Publication status	Published - 2021
Event	2021 IEEE Energy Conversion Congress and Exposition - Vancouver, Canada Duration: 10 Oct 2021 → 14 Oct 2021

Conference

Conference	2021 IEEE Energy Conversion Congress and Exposition
Country/Territory	Canada
City	Vancouver
Period	10/10/2021 → 14/10/2021

Keywords

Power Electronics
Reliability
Generative adversarial network
GAN
Mission profiles
Power converters

Access to Document

10.1109/ECCE47101.2021.9595614

OpenUrl availability

Full text

Cite this

@inproceedings{4f19ad3bcf8e40738286aa0b09e854fb,

title = "Employing the Generative Adversarial Networks (GAN) for Reliability Assessment of Converters",

abstract = "Mission profiles are widely used for the reliability analysis of power converters. Typically, to assess the converter reliability, long-term (e.g., one year) mission profiles are adopted, and it is assumed that the profiles will be repeated in future years. However, due to mission profile uncertainties, the assumption can introduce considerable errors in the estimated reliability. In this paper, the errors introduced by the above assumption are studied in detail. Furthermore, to tackle this challenge, the paper proposes using the Generative Adversarial Networks (GAN) to generate unique mission profile scenarios that capture the temporal and probabilistic properties of the real profiles. In this regard, the effectiveness of using the GAN-generated profiles to improve the accuracy of the estimated reliability is demonstrated.",

keywords = "Power Electronics, Reliability, Generative adversarial network, GAN, Mission profiles, Power converters",

author = "Amirali Davoodi and Saeed Peyghami and Yongheng Yang and Tomislav Dragicevic and Frede Blaabjerg",

year = "2021",

doi = "10.1109/ECCE47101.2021.9595614",

language = "English",

isbn = "978-1-7281-6128-0",

pages = "3623--3629",

booktitle = "Proceedings of 2021 IEEE Energy Conversion Congress and Exposition",

publisher = "IEEE",

address = "United States",

note = "2021 IEEE Energy Conversion Congress and Exposition, 2021 ECCE ; Conference date: 10-10-2021 Through 14-10-2021",

}

Davoodi, A, Peyghami, S, Yang, Y, Dragicevic, T & Blaabjerg, F 2021, Employing the Generative Adversarial Networks (GAN) for Reliability Assessment of Converters. in Proceedings of 2021 IEEE Energy Conversion Congress and Exposition. IEEE, pp. 3623-3629, 2021 IEEE Energy Conversion Congress and Exposition, Vancouver, British Columbia, Canada, 10/10/2021. https://doi.org/10.1109/ECCE47101.2021.9595614

TY - GEN

T1 - Employing the Generative Adversarial Networks (GAN) for Reliability Assessment of Converters

AU - Davoodi, Amirali

AU - Peyghami, Saeed

AU - Yang, Yongheng

AU - Dragicevic, Tomislav

AU - Blaabjerg, Frede

PY - 2021

Y1 - 2021

N2 - Mission profiles are widely used for the reliability analysis of power converters. Typically, to assess the converter reliability, long-term (e.g., one year) mission profiles are adopted, and it is assumed that the profiles will be repeated in future years. However, due to mission profile uncertainties, the assumption can introduce considerable errors in the estimated reliability. In this paper, the errors introduced by the above assumption are studied in detail. Furthermore, to tackle this challenge, the paper proposes using the Generative Adversarial Networks (GAN) to generate unique mission profile scenarios that capture the temporal and probabilistic properties of the real profiles. In this regard, the effectiveness of using the GAN-generated profiles to improve the accuracy of the estimated reliability is demonstrated.

AB - Mission profiles are widely used for the reliability analysis of power converters. Typically, to assess the converter reliability, long-term (e.g., one year) mission profiles are adopted, and it is assumed that the profiles will be repeated in future years. However, due to mission profile uncertainties, the assumption can introduce considerable errors in the estimated reliability. In this paper, the errors introduced by the above assumption are studied in detail. Furthermore, to tackle this challenge, the paper proposes using the Generative Adversarial Networks (GAN) to generate unique mission profile scenarios that capture the temporal and probabilistic properties of the real profiles. In this regard, the effectiveness of using the GAN-generated profiles to improve the accuracy of the estimated reliability is demonstrated.

KW - Power Electronics

KW - Reliability

KW - Generative adversarial network

KW - GAN

KW - Mission profiles

KW - Power converters

U2 - 10.1109/ECCE47101.2021.9595614

DO - 10.1109/ECCE47101.2021.9595614

M3 - Article in proceedings

SN - 978-1-7281-6128-0

SP - 3623

EP - 3629

BT - Proceedings of 2021 IEEE Energy Conversion Congress and Exposition

PB - IEEE

T2 - 2021 IEEE Energy Conversion Congress and Exposition

Y2 - 10 October 2021 through 14 October 2021

ER -

Employing the Generative Adversarial Networks (GAN) for Reliability Assessment of Converters

Abstract

Conference

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this