摘要
本研究应用大数据方法,挖掘信息化煤矿长期积累的海量数据,多维度洞察发现海量数据隐含的内在规律,将煤矿安全生产管理提升到精细化的新层面。本研究主要包括:收集大量煤矿安全事故报告,针对煤矿安全事故报告进行文本挖掘,在此基础上运用Delphi专家调查法确定事故关键要素;利用Apriori关联分析法分析关键要素之间的关联关系,挖掘关键要素频繁项集;依据煤矿现有信息化系统数据,建立与关键要素对应的数据指标并对其进行实时监控和预测;在关键要素频繁项集的基础上构建BP神经网络模型,通过计算实时数据预测事故发生的可能性,为煤矿事故的研究和预防提供了全新的视角。
This study applied the big data method to dig the long term accumulation of massive data in informatized coal mines. Multi-dimensional insights have been made to discover the inherent rules inherent in massive data,and to improve the safety management of coal mines to a new level of refinement. This study mainly includes：collecting a large number of coal mine safety accident reports and doing text mining,and the key elements of the accident were determined by using Delphi expert survey method. Using Apriori association analysis method to analyze the relationship between key elements and excavate frequent itemsets of key elements. According to the existing information system data of coal mines,data indicators corresponding to key elements are established and monitored and predicted in real time. A BP neural network model is constructed on the basis of frequent itemsets of key elements. And the possibility of accidents is predicted by calculating real time data,which provided a new perspective for the study and prevention of coal mine accidents.
作者
李东
周勇
Li Dong;Zhou Yong(China Shenhua Energy Co.,Ltd.,Beijing 100011,China)
出处
《煤炭经济研究》
2018年第6期39-45,共7页
Coal Economic Research
关键词
文本挖掘
事故要素
数据指标
神经网络
text mining
accident factors
data indicators
neural network
