基于BP神经网络的煤与瓦斯突出危险性预测的研究

【作者】 李春辉；

【导师】 陈日辉；

【作者基本信息】 昆明理工大学 ， 安全技术及工程， 2010， 硕士

【摘要】 煤矿安全是当前煤矿安全生产工作的首要任务,然而煤与瓦斯突出是威胁煤矿安全生产最为严重的自然灾害之一。它是一种极其复杂的动力现象,是地应力、瓦斯和煤的物理力学性质等因素综合作用的结果。鉴于目前突出灾害日益严重的现实状况,煤与瓦斯突出的预测研究对保障煤矿的安全生产显得更加重要。本文结合目前大家普遍接受的综合假说为理论基础,通过对贵州地区煤与瓦斯突出事故的统计分析,研究了煤与瓦斯突出的一般规律,并应用BP神经网络的知识对煤与瓦斯突出的危险性预测进行了研究。1)论文首先阐述了煤与瓦斯突出危险性预测的必要性和深远意义,分析了当前的研究现状和存在问题,并以综合假说的突出机理为基础,确定了基于BP神经网络的煤与瓦斯突出危险性预测的研究方案和技术路线。2)通过对该地区突出事故的统计分析,研究了煤与瓦斯突出的一般规律,并以综合假说为理论基础,分析了煤与瓦斯突出危险性的影响因素,建立了煤与瓦斯突出危险性的评价指标体系。3)分析以地应力、瓦斯参数和煤体物理力学性质等控制和影响突出的各种因素,应用三角模糊数的理论和知识,确定了煤与瓦斯突出危险性预测的主控因素。4)建立了基于BP神经网络的煤与瓦斯突出危险性预测的数学模型,并通过数学软件MATLAB将其实现,以响水矿井为例进行了区域突出危险程度的评价。该方法简单易行,具有很强的实用性和操作性。5)最后,提出了煤与瓦斯突出的区域性防治技术,即开采保护层、预抽煤层瓦斯和煤层注水等方法。并针对我国数字化矿山的建设需要,提出了基于GIS的煤与瓦斯突出危险性预测管理系统的设计方案,为煤与瓦斯突出的防治提供了种新的防治技术,开辟煤与瓦斯突出防治的新途径,对保障煤矿企业的安全生产具有重要的意义。本文通过对该地区突出的一般规律以及突出危险控制因素的研究,建立了基于BP神经网络煤与瓦斯突出危险性预测的数学模型,提出了煤与瓦斯突出区域性的防治技术,为煤与瓦斯突出的预测和防治探索新的途径。更多还原

【Abstract】 The safety of coal mine is the most important task in the current coal mine safety production, however, Coal and gas outburst is one of the most serious natural disasters in coal mine safety production. It is a kind of complicated dynamical phenomena in mines, and is the synthesized result of stress, gas, coal physics mechanics etc. because of the disasters increasingly heavier, the study of coal and gas outburst prediction is even more important for safety production in coal mine. Combined with the generally accepted synthesizing hypothesis as theory basis in the paper, the paper through the statistical analysis of the coal and gas outburst accident of Gui Zhou region, to study the general rules coal and gas outburst in the region’s, and the prediction of coal and gas outburst is studied by applying the knowledge of BP neural network.1)The need and significance of coal and gas outburst risk prediction have been discussed in the paper, the actualities and limitations of current studying have been summarized, the study project and technology routes of coal and gas outburst risk prediction also have been confirmed based on BP neural network, and the outburst mechanism of synthesizing hypothesis.2)The paper through the statistical analysis of the coal and gas outburst accident, to study the general rules coal and gas outburst in the region’s, analyzing the influencing factors of coal and gas outburst based on the theory basis of synthesizing hypothesis, a hazard assessment index system of coal and gas outburst was build.3)The paper analyzes a variety of factors controlling and effecting coal and gas outburst about the synthesized result of stress, gas, coal physics mechanics etc, the controlling factors of coal and gas outburst hazard prediction was confirmed by applying the theory and knowledge triangular fuzzy numbers.4)The mathematical model of coal and gas outburst hazard prediction was established based on BP neural network, it was achieved by mathematical software of MATLAB, for example of Xiang Shui mine, the risk degree of the region outburst was evaluated. The method is simple、very practical and operational.5)Finally, the regional control technology of coal and gas outburst was proposed, namely mining protective coal seam、gas pre-drainage and coal seam injecting etc. because of the need of digital mine, a design project of coal and gas outburst risk prediction management system based on GIS, a new control technology of coal and gas outburst was provided,a new access of coal and gas outburst control technology was opened, it have very important significance for safety production in mining enterprises.The paper through the research of the general rules outburst in the region’s and control factors of outburst hazard, the mathematical model of coal and gas outburst hazard prediction was established based on BP neural network, the regional control technology of coal and gas outburst was proposed, to explore new a avenue for prediction and control of coal and gas outburst.更多还原