Bayes判别模型在风化基岩富水性预测中的应用

(1.西安科技大学 地质与环境学院,陕西 西安 710054; 2.陕西省煤炭绿色开发地质保障重点实验室,陕西 西安 710054; 3.中煤能源研究院有限责任公司,陕西 西安 710054; 4.陕煤集团神木张家峁矿业有限公司,陕西 榆林 719316)

风化基岩; Bayes判别; 富水性; 张家峁井田

Application of Bayes discriminant model in prediction of water enrichment of weathered bedrock
HOU En-ke1,2,YAN Xin1,3,ZHENG Yong-fei4,YANG Fan4

(1.College of Geology and Environment,Xi'an University of Science and Technology,Xi'an 710054,China; 2.Shaanxi Provincial Key Laboratory of Geological Support for Coal Green Exploitation,Xi'an 710054,China; 3.China Coal Energy Research Institute Co.,Ltd.,Xi'an 710054,China; 4.Shenmu Zhangjiamao Mining Co.,Ltd.,Shaanxi Coal and Chemical Industry Group,Yulin 719316,China)

weathered bedrock; Bayes discrimination; water enrichment; Zhangjiamao coal mine

备注

风化基岩含水层是对矿井安全生产构成威胁的主要含水层之一,准确预测其富水性是煤矿水害防治的基础。针对风化基岩富水性预测问题,以张家峁井田为例,在总结风化基岩岩性特征、分布特征和富水性特征的基础上,以风化基岩厚度、风化程度、岩性组合指数、脆塑性岩厚度比和地形地貌作为判别指标,以研究区及其外围29组风化基岩抽水试验数据作为训练及检验样本,建立了风化基岩含水层富水性Bayes判别模型,可对无抽水试验资料区域的风化基岩富水性类型进行预测。研究表明,张家峁井田风化基岩整体富水性较弱,且空间分布不均。在井田中部和西北角局部区域为强富水性,中部大部分区域为中等富水性,东部及东北部区域几乎全为极弱富水性,其余部分为弱富水性。通过与矿井生产过程中井下涌水量观测数据对比,显示所建判别模型具有较高的准确率,表明运用此模型进行风化基岩富水性预测是可行的,为风化基岩含水层的富水性预测提供了一种新方法。

Weathered bedrock aquifer is one of the main aquifers that pose a threat to the safety of mine production.Accurate prediction of water enrichment is the basis of water disaster prevention and control coal mine.Aiming at the prediction of water enrichment of weathered bedrock,the Bayes discriminant model of weathered bedrock aquifer has been established with Zhangjiamao coal mine field area,to predict the water enrichment characteristics of weathered bedrock areas.In that process,a comprehensive analysis of the lithologic characteristics,distribution characteristics and water enrichment characteristics of weathered bedrock,thickness was made,and the thickness and weathered degree as well as lithology combination index,the ratio of thickness of brittle rock to plastic rock,topography and landform were chosen as discriminant indicators,taking the data of weathered bedrock from 29 group of pumping tests in the study area and its periphery as training and testing examples.Then,the model of water enrichmet of weathered bedrock aquifer based on Bayes discriminant has been built.The water enrichment type of weathered bedrock in drilling areas without pumping test data can be predicted.The result shows that the overall water enrichment of weathered bedrock in well field is weak and the spatial distribution is uneven.In the middle and northwest corner,the local area is strongly rich in water and the middle part moderately; the eastern and northeastern regions are very weak in water,and so are the rest parts are weak enrichment.A comparison the observed data of underground water inflow in the process of mine production shows that the established discriminant model is high in accuracy,indicating that it is feasible to use this model to predict the water enrichment of weathered bedrock aquifer.It provides a new method for predicting the water enrichment of weathered bedrock aquifer.