基于指标气体关联分析的煤自燃分级预警研究

为准确预警煤炭自然发火,基于程序升温试验,获得指标气体与温度间的关系,并将起始温度V0和特征温度点V1、V2、V3时刻的碳氧化物比率作为预警界限,设定4级预警机制,采用灰色关联分析法关联性分析采空区和上隅角CO气体体积分数和碳氧化物比率,按照关联等级的高低细化4级预警机制,同时选取典型案例,详细展示该预警机制下的预警流程。结果表明:时间序列点(133)CO体积分数的关联系数为0.4812,关联等级为D级,关联性差,而碳氧化物比率的关联系数为0.9975,关联等级为a级,关联性强,综合判断关联性较强(Da),发布蓝色Ⅱ级预警。

Research on grading and early warning of coal spontaneous combustion based on correlation analysis of index gas

In order to warn occurrence of coal spontaneous combustion accurately,relationship between index gas and temperature was obtained based on temperature-programmed test.Starting temperature V0 and carbon oxide ratio at characteristic temperature points V1,V2,and V3 were taked as warning limit,a 4-level early warning mechanism was set.Gray correlation analysis method was used to correlate CO gas volume fraction and carbon oxide ratio in the goaf and the upper corner,and 4-level early warning mechanism was refined according to level of correlation.At the same time,a typical case was selected and early warning process under early warning mechanism was shown in detail.The results show that correlation coefficient of the time series point(133)CO volume fraction is 0.4812,correlation level is D level and correlation is poor,while correlation coefficient of carbon oxide ratio is 0.9975,correlation level is a level and correlation is strong,comprehensive judgment is more relevant.Strong(Da),a blue levelⅡearly warning is issued.