An end-to-end power knowledge extraction method based on convolutional neural networks and multi-headed attention mechanisms

Bin Zhang; Yangjun Zhou; Liwen Qin; Shan Li

doi:10.1117/12.3005282

25 September 2023 An end-to-end power knowledge extraction method based on convolutional neural networks and multi-headed attention mechanisms

Bin Zhang, Yangjun Zhou, Liwen Qin, Shan Li

Author Affiliations +

Proceedings Volume 12788, Second International Conference on Energy, Power, and Electrical Technology (ICEPET 2023); 127884G (2023) https://doi.org/10.1117/12.3005282
Event: Second International Conference on Energy, Power, and Electrical Technology (ICEPET 2023), 2023, Kuala Lumpur, Malaysia

Abstract

The electric power energy industry, after decades of development, already possesses a large amount of technical literature in Chinese, which contains a wealth of expert knowledge. Intelligent decision-making would undoubtedly be faster and more accurate if knowledge could be accurately extracted and presented to employees in an understandable form or in an intelligent QA system. Facing the impact of massive electric text data, the technical problem addressed in this paper is to propose an end-to-end model based on graph convolutional neural networks and a multi-headed attention mechanism that combines contextual semantic features and syntactic features of text sequences to effectively improve the performance of the knowledge extraction task. Researching knowledge organisation methods for heterogeneous data in distribution network operation and maintenance, constructing expert knowledge-based extraction models and realising knowledge representation that flexibly and clearly expresses business logic can help improve the automation of power systems and provide global knowledge support for smart grids.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Bin Zhang, Yangjun Zhou, Liwen Qin, and Shan Li "An end-to-end power knowledge extraction method based on convolutional neural networks and multi-headed attention mechanisms", Proc. SPIE 12788, Second International Conference on Energy, Power, and Electrical Technology (ICEPET 2023), 127884G (25 September 2023); https://doi.org/10.1117/12.3005282

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