Research on deep-learning-based fault diagnosis and prediction methods for electrical systems

Wenhao Gu

doi:10.1117/12.3025418

28 February 2024 Research on deep-learning-based fault diagnosis and prediction methods for electrical systems

Wenhao Gu

Author Affiliations +

Proceedings Volume 13071, International Conference on Mechatronic Engineering and Artificial Intelligence (MEAI 2023); 1307121 (2024) https://doi.org/10.1117/12.3025418
Event: International Conference on Mechatronic Engineering and Artificial Intelligence (MEAI 2023), 2023, Shenyang, China

Abstract

With the breakthrough progress of deep learning technology in various fields, its application in fault diagnosis and prediction of electrical systems has received more and more attention. In this paper, a deep learning-based fault diagnosis and prediction model is proposed for the complex nonlinear characteristics in electrical systems. First, key features are automatically extracted from a large amount of electrical system operation data using a multilayer autoencoder (MLAE). These features are fed into a deep neural network (DNN) for fault classification and prediction. In order to improve the robustness and accuracy of the model, an attention mechanism is introduced so that the model pays more attention to the key features related to faults during the learning process. The method demonstrates high accuracy and reliability on multiple electrical system fault datasets compared to traditional electrical system fault diagnosis methods. In addition, this study explores the interpretability.

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

Citation Download Citation

Wenhao Gu "Research on deep-learning-based fault diagnosis and prediction methods for electrical systems", Proc. SPIE 13071, International Conference on Mechatronic Engineering and Artificial Intelligence (MEAI 2023), 1307121 (28 February 2024); https://doi.org/10.1117/12.3025418

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