【作者】 杨国梁；

【导师】 吴世跃；

【作者基本信息】 太原理工大学 ， 安全技术及工程， 2010， 硕士

【摘要】 煤与瓦斯突出是极其复杂的一种动力现象,是严重威胁矿山安全生产的一种自然灾害。煤与瓦斯突出预测的关键是在众多突出预测指标中的筛选出适合要预测煤矿的敏感指标及其临界值的确定。现常用的方法是“三率”法和灰色关联分析法。其中“三率”法对井下实测数据中的突出动力现象进行详细的记录,极大程度的依靠工作人员的经验,误差较大。而灰色关联分析法数据处理过程复杂,不适合在现场生产中推广应用。针对以上突出预测敏感指标确定方法的不足,本文作了如下研究：1、结合煤与瓦斯突出机理及发生条件综合分析了现行的突出预测指标的应用现状、测定方法和各指标所反映的突出因素。2、对主成分分析法在煤与瓦斯突出预测指标敏感性排序中的应用进行了尝试。通过主成分分析法原理的分析,结合突出预测指标的特点,论述了主成分分析法在煤与瓦斯突出预测指标敏感性排序中应用的可行性。对某煤矿的实测数据同时应用“三率”法,灰色关联分析法和主成分分析法进行了突出预测指标敏感性排序。三种方法得出的结论一致。证明了主成分分析法在突出预测指标敏感性排序中的可靠性。3、针对井田范围大,地质条件复杂的大型煤矿提出了先根据突出预测指标所反映的突出因素对矿井进行区域划分,然后确定不同的区域临界值的方法。以寺河煤矿为例,针对钻屑瓦斯解吸指标对寺河煤矿进行区域划分。选择煤的坚固性系数变化、软煤层带、瓦斯含量变化等因素作为区域划分的标准。根据划分标准将煤矿划分出寺河东、西两大区,通过实验室方法分别得出适合寺河煤矿西区的钻屑瓦斯解吸指标K1试验临界值为0.83ml／(min0.5·g),东区为0.75 ml／(min0.5·g)。然后进行现场测量,以预测突出准确率和预测不突出准确率为可靠性判定指标进行临界值优化。通过这样的步骤进行临界值确定,能够使寺河煤矿减少了大约1／3的防突工作,占总预测总数的9%。临界值的确定周期大大缩短,提高了临界值优化阶段的工程效率,同时使钻屑解吸指标的临界值有更强的针对性。论文最大的创新点是首次将主成分分析法应用在煤与瓦斯突出预测指标敏感性排序中,提出了对地质条件复杂的大型煤矿进行突出预测时应分区域确定不同临界值的思路,并论述了针对钻屑瓦斯解吸指标对大型煤矿进行区域划分的原则。更多还原

【Abstract】 Coal and gas outburst is a complacate dynamic phenomenon, and a natural calamity which seriously threaten the mine production security. The key of coal and gas outburst forecast is screening out the right sensitive index from so many outburst forecast indexes,commonly used current method is the three rates rule and gray relational analysis.Three rates rule’s process is clear and easy to understand, and calls for the outburst dynamic phenomena record in detail,and the determining of the outburst dynamic phenomena depend on the experience of the workers,mistake is usual.Although the gray relational analysis method avoid a three rates rule’s fuzzy determination of the dynamic phenomena, but the data processing complex, not suitable for application in the field to promote the production. Aimed at the above lack of determining sensitive indexes,reasch is done as follow:1、Base on coal and gas outburst mechanism and occur conditions,this paper comprehensivly analyze the current application status, measurement methods and the prominent factor reflected by indexes. 2、Have tried on using Principtal Component Analysis on screening of coal and gas outburst prediction index. Through analysising principal component theory, combined with the characteristics of outburst predict index, discussed the feasibility of the principal component analysis used in the coal and gas outburst prediction index screening. Analysis a batch of measured data of a coal mine by three rates rule,gray correlation analysis and principal component analysis were prominent predictor of sensitivity ranking, the conclusions of three methods is consistent. This prove the reliablity of Principal component analysis used on demonstrated prominent predictor of sensitivity.3、For the coal mine which has lardge range and complex geological conditions, this paper suggest that the coal ming shoule be divided into several small range according to the outburst factor reflected by outburst proforst indexes, and then confirm different critical values of the desorption index for regions. In this paper divided sihe coal mine into two regions according to gas content, consistent coefficient. Confirm the critical value of desorption index for drill cuttings for one region by laboratory method. Get the result as 0.83 ml/ (min0.5·g) and 0.75 ml/(min0.5·g)Though this way, reduce the out burst work by 1/3,which account 9% of the whole outburst control, and shortened the cycle of confirming the critical value on field.The innovation of this paper is the first time tried use principal component analysis on screening coal and gas outburst prediction sensitivity indexes, and analysis the principle of confirming of the critical value of the desorption index for regions of big coal mine with complicated geological condition.更多还原