E3S Web Conf.
Volume 182, 20202020 10th International Conference on Power, Energy and Electrical Engineering (CPEEE 2020)
|Number of page(s)||5|
|Section||Modern Power System Control and Operation|
|Published online||31 July 2020|
Application of natural language understanding in Chinese power dispatching centre
1 School of Electrical Engineering, South China University of technology, Guangzhou 510640, China
2 Power Dispatching and Controlling Center, Guangzhou Power Supply Co., Ltd., Guangzhou 510620, China
3 Reading Academy, Nanjing University of Information Science and Technology, Nanjing 210044, China
* Corresponding author: email@example.com
It is difficult for computer to understand the texts in unstructured Chinese language, which becomes an obstacle for further application of artificial intelligence in the power dispatch center. Understanding of the orders from human dispatchers is the premise for the collaboration of machine and human being in power system operation. Towards understanding of dispatching texts, this paper proposes a textual semantic analysis framework with active learning of the semantic structure knowledge. Firstly, the words are vectorized by the Skip-gram models. And the hierarchical clustering algorithm is designed to detect the sentence patterns. Then the knowledge base is set up by converting the sentence structure to their regular expressions. In application, define a proprietary semantic framework to extract important device information and to parse the semantic slot using dependency syntax. Application shows that the Chinese texts describing the operation mode switching process can be understood accurately by the computer program.
© The Authors, published by EDP Sciences, 2020
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.