煤矿海量通风安全数据挖掘算法及智能分析系统研究

发布时间：2018-03-10 04:30

本文选题：数据挖掘　切入点：智能分析　出处：《太原理工大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着信息化建设的逐渐深入,煤炭企业的管理决策部门在前期信息化建设中的不足和薄弱环节逐渐暴露出来。弥补这一不足的紧迫性逐渐提上日程,我们有必要研发一套符合实际情况的,能够替代现有人工数据整理工作的,快捷高效的智能化数据分析系统。由于来自基层的原始数据的不断日积月累,生产运行数据增长的特别快,海量数据成为当今一个趋势,时至今日,采取一定方法分析数据提取信息很重要,提取有用的信息更重要,这便要求一套面向安全的,能够实现数据整合的并迅速得到有价值信息的,具有海量数据处理能力的,支持数据挖掘的智能数据分析系统。本文在研究过程中,结合相关理论研究、算法模型,主要工作包括以下几个方面:(1)首先根据数据挖掘理论基础,研究总结数据挖掘常用技术与算法,提出了数据挖据在用于通风安全灾害分析的三部曲:就是数据准备、数据挖掘与结果评估,为数据挖掘技术用于智能分析系统奠定基础。(2)介绍了基于煤矿海量通风数据进行数据挖掘的算法模型,重点介绍MGM(1,n)—BP神经网络通风安全数据序列研究模型,并且在实际工程应用中得到验证。(3)本着智能分析技术处理海量数据的方法,从煤矿数据智能分析系统的体系结构、智能分析系统ETL、煤矿数据智能分析系统处理过程三个方面介绍了煤矿智能分析系统。(4)整个系统建立在现有各类原始数据基础之上,实现对原始数据的提炼和深加工,最终为通风安全工作提供有价值的指导性信息。(5)系统信息展示层以多种形式将智能分析得到的有价值信息进行展示,实现分析结果的可视化。这其中包括使用各种图形和表格(与现在手工模式的Excel格式兼容)以及动画对数据进行展示。
[Abstract]:With the gradual deepening of information construction, the shortcomings and weaknesses of the management and decision-making departments of coal enterprises in the early stage of information construction are gradually exposed. The urgency of making up for this deficiency is gradually put on the agenda. It is necessary for us to develop a fast, efficient and intelligent data analysis system that conforms to the actual situation and can replace the existing manual data collation. Because of the accumulation of raw data from the grassroots, Production and operation data are growing very fast, mass data has become a trend nowadays, it is very important to take a certain method to analyze data to extract information, and it is more important to extract useful information, which requires a set of security-oriented, Intelligent data analysis system, which can realize data integration and get valuable information quickly, has the ability of massive data processing, and supports data mining. The main work includes the following aspects: (1) first of all, according to the theoretical basis of data mining, the common techniques and algorithms of data mining are studied and summarized, and the trilogy of data mining for ventilation safety disaster analysis is put forward: data preparation. This paper introduces the algorithm model of data mining based on massive ventilation data in coal mine, and introduces the research model of ventilation safety data sequence based on MGMM1 / NN-BP neural network, which lays a foundation for the application of data mining and result evaluation in intelligent analysis system. And it has been verified in the practical engineering application. (3) based on the method of intelligent analysis technology to deal with massive data, from the system structure of intelligent analysis system of coal mine data, The intelligent analysis system ETL, the processing process of coal mine data intelligent analysis system is introduced. The whole system is built on the basis of all kinds of existing raw data, and realizes the extraction and deep processing of the original data. Finally, it provides valuable guidance information for ventilation safety work. The system information display layer displays the valuable information obtained by intelligent analysis in various forms. Visualization of analysis results. This includes the use of various graphics and tables (compatible with the Excel format of today's manual mode) and animation to display the data.
【学位授予单位】：太原理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TD724

【参考文献】