基于语义特征抓取电网调度事件的检测技术

doi:10.16183/j.cnki.jsjtu.2021.S2.014

上海交通大学学报 ›› 2021, Vol. 55 ›› Issue (S2): 86-91.doi: 10.16183/j.cnki.jsjtu.2021.S2.014

基于语义特征抓取电网调度事件的检测技术

许凌¹, 王兴志¹, 肖林朋²^,³()

1.国家电网有限公司华东分部,上海 200120
2.南瑞集团有限公司(国网电力科学研究院有限公司),南京 211106
3.北京科东电力控制系统有限责任公司,北京100192

收稿日期:2021-09-01 出版日期:2021-12-28 发布日期:2022-01-24
通讯作者: 肖林朋 E-mail:1553990434@qq.com
作者简介:许凌(1981-),男,山东省平度市人,高级工程师,从事大电网调度运行及自动控制研究.
基金资助:
国家电网有限公司华东分部科技项目(SGHD0000DKJS2100225)

Power Grid Dispatching Event Detection Technology Based on Semantic Feature Capture

XU Ling¹, WANG Xingzhi¹, XIAO Linpeng²^,³()

1. East Branch of State Grid Corporation of China, Shangha 200120, China
2. Nari Group Corporation (State Grid Electric Power Research Institute), Nanjing 211106, China
3. Beijing Kedong Power Control System Co., Ltd., Beijing 100192, China

Received:2021-09-01 Online:2021-12-28 Published:2022-01-24
Contact: XIAO Linpeng E-mail:1553990434@qq.com

摘要/Abstract

摘要：

电网运行管理文档中的领域用语专业且复杂,在信息化提取的过程中需要优秀且适用的事件检测方法抽取事件主体.而当前的中文事件检测方法大多采用词嵌入技术以捕获语义表示,但这些方法难以捕捉触发词与同一句子中的其他领域词汇间的依赖关系.基于上述现状,提出了一种混合表示的新架构,以表征两个字符和单词的语义和结构信息.该模型可以通过依赖性解析器生成的语义表示来捕获富语义特征.实验基于调度日志、调度检修票、调度预案语料使用不同模型进行评估,结果表明,该方法可以显著提高电网调度文本事件检测的性能.

关键词: 电网调度, 事件检测, 神经网络, 语义特征

Abstract:

The domain terms in power grid operation management documents are professional and complex. In the process of information extraction, excellent and applicable event detection methods are needed to extract event subjects. However, most of the current Chinese event detection methods use the word embedding technology to capture semantic representation, but it is difficult for these methods to capture the dependency between trigger words and other domain words in the same sentence. Based on the above situation, this paper proposes a novel hybrid representation architecture to represent the semantic and structural information of two characters and words. The model can capture rich semantic features through the semantic representation generated by the dependency parser. The experiment is based on the corpus of dispatching log, dispatching maintenance ticket, and dispatching plan. The results show that this method can significantly improve the performance of power grid dispatching text event detection.

Key words: power grid dispatching, event detection, neural network, semantic feature

中图分类号:

TP311

许凌, 王兴志, 肖林朋. 基于语义特征抓取电网调度事件的检测技术[J]. 上海交通大学学报, 2021, 55(S2): 86-91.

XU Ling, WANG Xingzhi, XIAO Linpeng. Power Grid Dispatching Event Detection Technology Based on Semantic Feature Capture[J]. Journal of Shanghai Jiao Tong University, 2021, 55(S2): 86-91.

图/表 5

表1

图1

图2

图3

表2

参考文献 8

[1]	ZHANG T T, JI H, SIL A. Joint entity and event extraction with generative adversarial imitation learning[J]. Data Intelligence, 2019, 1(2): 99-120. doi: 10.1162/dint_a_00014 URL
[2]	LIU S L, CHEN Y B, LIU K, et al. Exploiting argument information to improve event detection via supervised attention mechanisms[C]// 55th Annual Meeting of the Association for Computational Linguistics. Vancouver: ACL, 2017: 1789-1798.
[3]	YANG Y M, PIERCE T, CARBONELL J. A study of retrospective and on-line event detection[C]// 21st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. Melbourne: ACM, 1998: 28-36.
[4]	YANG Y M, ZHANG J, CARBONELL J, et al. Topic-conditioned novelty detection[C]// Eighth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Edmonton: ACM, 2002: 688-693.
[5]	SALTON G. Automatic text processing: The transformation analysis and retrieval of information by computer[M]. Reading: Addison-Wesley, 1989.
[6]	KUMARAN G, ALLAN J. Text classification and named entities for new event detection[C]// 27th Annual International Conference on Research and Development in Information Retrieval. Sheffield: ACM, 2004: 297-304.
[7]	SANKARANARAYANAN J, SAMET H, TEITLER B E, et al. TwitterStand: News in tweets[C]// 17th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. Seattle: ACM, 2009: 42-51.
[8]	PHUVIPADAWAT S, MURATA T. Breaking news detection and tracking in twitter[C]// 2010 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology. Toronto: IEEE, 2010: 120-123.

编号	句子	触发词
1	A省火电站总体装机容量为x kW	电站类型
2	供电员工站立时的行为举止规范要求	身体姿态

模型	触发识别			触发分类
模型	P	R	F1	P	R	F1
FBRNN	62.23	37.8	45.98	51.87	32.78	40.82
DMCNN	56.27	23.82	46.92	67.21	32.98	45.98
CLIZH	63.21	23.45	33.72	60.76	35.21	36.87
NPN	65.87	66.78	33.87	56.98	45.89	40.30
JMCEE	67.28	67.89	36.89	49.90	35.89	40.37
本文方法	69.34	59.34	49.38	68.21	46.98	49.90

[1]	李诗杰, 刘泰序, 刘佳仑, 董智霖, 徐诚祺. 基于GRU-MPC的双全回转推进拖轮轨迹跟踪控制[J]. 上海交通大学学报, 2026, 60(3): 418-426.
[2]	梁佳铭, 余音, 胡祎乐. 基于时空神经网络的金属疲劳裂纹扩展预测[J]. 上海交通大学学报, 2026, 60(3): 511-521.
[3]	. DSNet：用于语音情感识别的带有中性校准的解耦孪生网络[J]. J Shanghai Jiaotong Univ Sci, 2026, 31(2): 248-257.
[4]	. 类间隙滞后非线性系统复合双通道干扰估计自适应控制器设计[J]. J Shanghai Jiaotong Univ Sci, 2026, 31(1): 106-116.
[5]	夏筱彦, 张宇, 胡锡坤, 钟平. 基于扩散模型的无人机遥感目标检测物理对抗攻击方法研究[J]. 空天防御, 2026, 9(1): 52-62.
[6]	李湘, 陈思远, 张俊, 柯德平, 高杰迈, 杨欢欢. 基于物理信息嵌入的非固定长度电力系统暂态稳定快速评估[J]. 上海交通大学学报, 2025, 59(7): 962-970.
[7]	王晓倩, 周羽生, 毛源军, 李彬, 周文晴, 苏盛. 基于神经网络分位数的分布式光伏发电功率异常识别方法[J]. 上海交通大学学报, 2025, 59(6): 836-844.
[8]	曾金灿, 何耿生, 李姚旺, 杜尔顺, 张宁, 朱浩骏. 基于卷积神经网络与轻量级梯度提升树组合模型的电力行业短期以电折碳方法[J]. 上海交通大学学报, 2025, 59(6): 746-757.
[9]	. 基于深度学习序列方法的多人姿态估计用来检测人体与关键点位置[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(6): 1103-1113.
[10]	张理, 王宝, 贾健雄, 宋竹萌, 叶钰童, 余跃, 林嘉庆, 徐潇源. 微电网功率预测与调度端到端协同优化方法[J]. 上海交通大学学报, 2025, 59(6): 720-731.
[11]	潘美琪, 贺兴. 基于零样本学习的风力机故障诊断方法[J]. 上海交通大学学报, 2025, 59(5): 561-568.
[12]	黄逸翔, 窦迅, 李林溪, 杨函煜, 于建成, 霍现旭. 基于全局灵敏度分析的综合能源设备响应价值量化方法[J]. 上海交通大学学报, 2025, 59(5): 569-579.
[13]	. 基于基尼不纯度结构优化物理引导神经网络的薄膜型声学超材料传声损失预测[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(3): 613-624.
[14]	. 基于毫米波雷达的智能心率提取方法[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(3): 493-498.
[15]	. 基于表面肌电信号的BP-LSTM混合模型肘部运动实时预测[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(3): 455-462.

基于语义特征抓取电网调度事件的检测技术

Power Grid Dispatching Event Detection Technology Based on Semantic Feature Capture

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 5

参考文献 8

相关文章 15

编辑推荐

Metrics

本文评价