Transformer fault diagnosis method based on ATT-CNN-Bi-LSTM

Kai Zhou; Jinghan Wu; Yulei Li; Haoyang Tang; Jingyi Yang; Zhijian Hu

doi:10.1117/12.3024682

13 May 2024 Transformer fault diagnosis method based on ATT-CNN-Bi-LSTM

Kai Zhou, Jinghan Wu, Yulei Li, Haoyang Tang, Jingyi Yang, Zhijian Hu

Proceedings Volume 13159, Eighth International Conference on Energy System, Electricity, and Power (ESEP 2023); 131596Q (2024) https://doi.org/10.1117/12.3024682
Event: Eighth International Conference on Energy System, Electricity and Power (ESEP 2023), 2023, Wuhan, China

Abstract

With the advancements in online monitoring technology for power equipment, real-time monitoring of transformers has become feasible. It is of immense significance to assess and predict the operational status of transformers in real-time. This paper introduces a methodology that utilizes ATT-CNN-Bi-LSTM for diagnosing the state of transformers. The proposed approach involves preprocessing transformer monitoring data and historical operation and maintenance data acquired from the central control station. It conducts initial feature extraction on fault data and extends overall state features from local state features using Bi-LSTM to achieve transformer state assessment. Through experimental comparative analysis, the method put forward in this paper effectively addresses the limitations of traditional deep learning methods, such as dependence on extensive data and low accuracy, which significantly enhances the predictive capability of transformer fault diagnosis.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Kai Zhou, Jinghan Wu, Yulei Li, Haoyang Tang, Jingyi Yang, and Zhijian Hu "Transformer fault diagnosis method based on ATT-CNN-Bi-LSTM", Proc. SPIE 13159, Eighth International Conference on Energy System, Electricity, and Power (ESEP 2023), 131596Q (13 May 2024); https://doi.org/10.1117/12.3024682

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