钱钢,金鑫,张锋明,朱峰,陈楠,陈武军,陈明强,汪力
QIAN Gang,JIN Xin,ZHANG Fengming,ZHU Feng,CHEN Nan,CHEN Wujun,CHEN Mingqiang,WANG Li

• 1:国网浙江省电力有限公司绍兴供电公司
• 2:宁波天灵信息科技有限公司

摘要(Abstract)：

提出基于自然语言处理的故障应急事故报告自动生成方法:采集故障设备相关数据存储在数据库内,利用得分匹配法对数据进行缺失值借补处理;采用自然语言处理技术进行资料分析,将复合文档内的数据信息划分为单词与句子两个数据结构,对数据结构进行词性标注与句法分析,并利用一致度计算进行语义识别;依照资料分析结果与模板需求自动进行监测数据分析报告配置与生成。实验结果显示,该方法自动生成故障应急事故报告过程中,数据借补、词性标注与语义识别等过程具有较高精度,可有效规范报告编制内容。
The paper introduces an automation generation method of fault emergency report based on natural language processing(NLP): collect the relevant data of the faulty equipment and store it in the database, and use the score matching method to perform the missing value compensation processing on the data; use NLP technology for data analysis, and divide the data information in the compound document into two data structures of words and sentences; perform part-of-speech tagging and syntactic analysis on the data structure, and use the consistency calculation for semantic recognition; configure and generate monitoring data analysis reports according to the data analysis results and template requirements. Experimental results show that in the process of automatically generating fault emergency reports, the process of data borrowing, part-of-speech tagging and semantic recognition has high accuracy, which can effectively standardize the content of the report.

关键词(KeyWords)： 自然语言;事故报告;自动生成;数据检验;词性标注;语义识别
natural language;emergency report;automatic generation;data inspection;part-of-speech tagging;semantic recognition

Abstract:

Keywords:

基金项目(Foundation): 国网浙江省电力有限公司科技项目(5211SX1900PT)

作者(Author): 钱钢,金鑫,张锋明,朱峰,陈楠,陈武军,陈明强,汪力
QIAN Gang,JIN Xin,ZHANG Fengming,ZHU Feng,CHEN Nan,CHEN Wujun,CHEN Mingqiang,WANG Li

DOI: 10.19585/j.zjdl.202103013

参考文献(References)：

• [1]袁长峰,陶彬,王万雷,等.油储系统火灾事故应急过程本质安全的分类风险源识别方法研究[J].工业安全与环保,2017,43(11):49-53.
• [2]胡林,易平,黄晶,等.基于真实事故案例的自动紧急制动系统两轮车测试场景研究[J].汽车工程,2018,40(12):1435-1446.
• [3]杨丽,张彤彤,周文杰.共词分析识别研究热点的效标关联效度研究：基于自然语言处理[J].图书与情报,2018(1):15-19.
• [4]周启,谭界雄,高全,等.大坝安全监测资料整编分析报告自动生成系统[J].人民长江,2019,50(1):215-219.
• [5]杨梦,周恩波.基于专家系统的煤矿事故现场处置方案自动生成系统研究[J].煤炭工程,2019,51(11):138-142.
• [6]阚洪海,赵杰.基于水晶报表的查收查引报告自动生成的设计与实现[J].现代情报,2017,37(4):129-133.
• [7]ZHANG X Y,KIM J,PATZER R E,et al.Prediction of emergency department hospital admission based on natural language processing and neural Networks[J].Methods Information in Medicin,2017,56(5):377-389.
• [8]赵呈领,胡萍,梁云真,等.在线开放课程中教师教学行为研究———结合自然语言处理观点挖掘的方法[J].中国远程教育,2019(1):58-66.
• [9]陈晨,王愉茜,欧阳志友,等.基于自然语言处理的蛋白质小分子亲和力值预测[J].应用科学学报,2019,37(3):327-335.
• [10]吴小坤,赵甜芳.自然语言处理技术在社会传播学中的应用研究和前景展望[J].计算机科学,2020,47(6):184-193.
• [11]徐琳宏,丁堃,林原,等.基于机器学习算法的引文情感自动识别研究———以自然语言处理领域为例[J].现代情报,2020,40(1):35-40.
• [12]白杨,张丽萍.挖掘软件源代码的代码注释自动生成方法[J].计算机工程与应用,2020,56(10):246-253.
• [13]WEI D W,WANG B,LIN G,et al.Research on unstructured text data mining and fault classification based on RNN-LSTM with malfunction inspection report[J].Energies,2017,10(3):406.
• [14]熊志恒,闵华松.基于自然语言的分拣机器人解析器技术研究[J].计算机工程与应用,2017,53(8):113-119.
• [15]BAI H,YANG S.Compact prediction tree(CPT)application research on the causes of faults in the use of multimedia classrooms in colleges and universities[J].Journal of Physics:Conference Series,2020,1453(1):120.