[1]崔铁军,李莎莎.SFEP文本因果关系提取及其与SFN转化研究[J].智能系统学报,2020,15(5):998-1005.[doi:10.11992/tis.201907021]
CUI Tiejun,LI Shasha.Causality extraction of SFEP text and its conversion to SFN[J].CAAI Transactions on Intelligent Systems,2020,15(5):998-1005.[doi:10.11992/tis.201907021]
点击复制
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
15
期数:
2020年第5期
页码:
998-1005
栏目:
学术论文—知识工程
出版日期:
2020-09-05
- Title:
-
Causality extraction of SFEP text and its conversion to SFN
- 作者:
-
崔铁军1,2, 李莎莎2
-
1. 辽宁工程技术大学 安全科学与工程学院,辽宁 阜新 123000;
2. 辽宁工程技术大学 工商管理学院,辽宁 葫芦岛 125105
- Author(s):
-
CUI Tiejun1,2, LI Shasha2
-
1. College of Safety Science and Engineering, Liaoning Technical University, Fuxin 123000, China;
2. School of Business Administration, Liaoning Technical University, Huludao 125105, China
-
- 关键词:
-
安全系统工程; 智能科学; 知识提取; 系统故障; 演化过程; 文本因果关系; 空间故障网络; 转化研究
- Keywords:
-
safety system engineering; intelligence science; knowledge extraction; system fault; evolution process; text causality; space fault network; transformation research
- 分类号:
-
X913; C931.1;TP18
- DOI:
-
10.11992/tis.201907021
- 文献标志码:
-
A
- 摘要:
-
为将系统故障演化过程(system fault evolution process,SFEP)的文本描述转化为空间故障网络(space fault network,SFN)结构,用于故障分析,本文提出SFEP文本因果关系提取方法,及其与SFN基本结构的转化方法。首先给出SFEP中事件的几种典型因果关系。随后提出因果关系与SFN基本结构的转化流程。本文方法围绕着关键字和因果关系组模式展开,通过模型的不断学习补充和丰富关键字和组模式。最终使方法具备将SFEP文本转化为SFN结构的能力。以飞机起落架故障发生过程文本为例进行了应用,实验结果表明该方法可用于SFEP文本中的因果关系分析,并得到了理想的SFN。完善的关键字和组模式有利于使用计算机智能处理SFEP的SFN。
- Abstract:
-
To transform the text description of the system fault evolution process (SFEP) into a space fault network (SFN) for fault analyses, we propose a method for extracting the text causality of SFEP and its transformation with the basic structure of SFN. First, several typical causal relationships of events in SFEP are given. Then, the transformation process of causality and SFN basic structure is proposed. The method focuses on the key words and causal group models; the keywords and group models are enriched and supplemented through continuous model learning. Finally, the method transforms SFEP text into an SFN structure. The method is applied to the text processing of aircraft landing gear failure. The findings show that the method can be used to analyze causality in SFEP text and obtain an ideal SFN. Perfect keywords and group models are beneficial to the intelligent processing of SFEP text to SFN using computers.
备注/Memo
收稿日期:2017-07-12。
基金项目:国家自然科学基金项目(52004120,51704141);国家重点研发计划项目(2017YFC1503102)
作者简介:崔铁军,副教授,主要研究方向为系统可靠性及力学系统稳定性。提出和建立了空间故障树理论及空间故障网络理论,获得多个期刊优秀论文奖。授权发明专利20项。发表学术论文100余篇,出版学术专著5部;李莎莎,讲师,主要研究方向为系统可靠性及安全管理。授权发明专利多项,获得多项省级期刊优秀论文奖。发表学术论文10余篇,出版学术专著3部
通讯作者:崔铁军.E-mail:ctj.159@163.com
更新日期/Last Update:
2021-01-15