距离判别法在相似矿区突水水源识别中的应用

Application of distance discriminance in identifying water inrush resource in similar coalmine

为了进行矿井突水水源预测,选用6种常规离子作为判别因子,并结合河南焦作矿区实测数据,建立了煤矿突水水源预测的距离判别分析模型;同时针对该模型自身检验的不足,采用灰色系统关联度法检验距离判别分析模型的准确性;将距离判别分析模型和灰色系统关联度法检验应用于华北型煤田水文地质条件相似的新安矿区,开展关联度分析。结果表明:选用6种常规离子浓度作为判别因子,可解决单位不一致的问题;选用灰色关联度法检验距离判别分析模型的正确性,保证了所选样品的充分使用,能很好地弥补两种方法的不足;距离判别分析模型排除了一般方法评分时存在的人为因素影响;各水样对参考序列的关联度在0.8左右,两矿区石炭系含水层水源具有较强的关联度。

The distance discriminant analysis model was established to predict the mine water inrushing sources effectively on the base of Jiaozuo mine area measured data by selecting six discrimination factors.The accuracy of distance discriminant analysis was checked by adopting gray related degree method.The distance discriminant analysis model and gray related degree method were used to Xin＇an mine area which hydrochemical character was similar to North-China type coalfields,and the gray related degree was also analyzed.The results show that the discrimination factors of six conventional ion concentration can solve inconsistency of unit.The gray related degree method can make sure that the selection of samples are been fully used and can make up for the two methods＇ deficiencies.The man-made influence factors can also be excluded by using the distance discriminant analysis model and the association degree of water samples can reach approximately 0.8.The carboniferous system aquifer of two mining areas have strong related degree.