基于多层关联规则算法的煤矿突出预警模型

为提升煤矿突出预警精度，提出基于多层关联规则算法的煤矿突出预警模型。分析造成煤矿突出问题产生的主要因素，并以此为基础，由地质构造、煤层赋存参数与瓦斯参数3方面出发，选取压扭性断层、煤厚变化量以及煤层瓦斯含量等煤矿突出预警指标，构建煤矿突出预警指标体系。采用层次分析法构建煤矿突出预警模型，利用多层关联规则算法挖掘模型内同层指标关联规则与跨层指标的关联规则，确定预警模型内各指标的权重，根据组合权重确定煤矿突出预警结果，并将预警结果分为安全、威胁、危险与突出4个等级。实验结果显示，所研究模型的预警精度基本保持在99%以上。

Coal mine outburst early warning model based on multi-layer association rule algorithm

In order to improve the accuracy of coal mine outburst warning,a coal mine outburst warning model using multi-layer association rule algorithm was proposed.The main factors causing coal mine outburst were analyzed. Based on this，the coal mine outburst early warning indicators such as fault，coal thickness change and coal seam gas content were selected from the three aspects of geological structure，coal seam occurrence parameters and gas parameters to build a coal mine outburst early warning indicator system.Early warning model of coal mine outburst was built by using analytic hierarchy process，the multi-layer association rule algorithm ws used to mine the association rules of the same layer index and the cross-layer index in the model，the weight of each index in the early warning model was determined; according to the combined weight，the coal outburst early warning result was determined.According to the combination weight，the early warning results of coal mine outburst was determined.And the early warning results were divided into four levels:safety,threat,danger and outstanding.Experimental results show that the early warning accuracy of the studied model basically remains above 99%.