Partial discharge fault diagnosis based on optimized BI-LSTM

Shutao Zhou; Yufan Wang

doi:10.1117/12.2639751

19 October 2022 Partial discharge fault diagnosis based on optimized BI-LSTM

Shutao Zhou, Yufan Wang

Proceedings Volume 12294, 7th International Symposium on Advances in Electrical, Electronics, and Computer Engineering; 122940J (2022) https://doi.org/10.1117/12.2639751
Event: 7th International Symposium on Advances in Electrical, Electronics and Computer Engineering (ISAEECE 2022), 2022, Xishuangbanna, China

Abstract

Long-term partial discharge (PD) can lead to equipment insulation failure, resulting in huge economic losses, and PD signals are difficult to capture. In order to improve the reliability of partial discharge, this paper proposes a PD classification algorithm based on bidirectional long short-term memory network and attention mechanism (BI-LSTM-Attention). This method takes the PD signal as input, uses the BI-LSTM network to extract the features of the sequence before and after. Combined with the attention mechanism to adjust the feature weight, and finally through the sigmoid activation function to classify the results into two categories, get the PD recognition results. Tested on the ENET public dataset, the experimental results show that compared with the traditional long short-term memory network (LSTM), the loss rate and accuracy rate of the classification results are better, indicating that the proposed model based on BI-LSTM-Attention is more effective for classification tasks, which reflects the superiority of the algorithm.

Citation Download Citation

Shutao Zhou and Yufan Wang "Partial discharge fault diagnosis based on optimized BI-LSTM", Proc. SPIE 12294, 7th International Symposium on Advances in Electrical, Electronics, and Computer Engineering, 122940J (19 October 2022); https://doi.org/10.1117/12.2639751

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Data modeling

Neural networks

Feature extraction

Performance modeling

Reliability

Signal attenuation

Dielectrics

Show All Keywords

Keywords/Phrases

Search In:

Publication Years