摘要
为了快速准确区分矿井涌水的来源,以柠条塔煤矿为例,通过对萨拉乌苏组含水层、直罗组风化基岩含水层、烧变岩含水层以及采空区积水进行水质化验,分析了不同含水层的水化学特征,选取Na^(+)+K^(+),Ca^(2+),Mg^(2+),Cl^(-),SO_(4)^(2-),HCO_(3)^(-),TDS的浓度作为水源识别的判别指标;利用逐步回归分析(SR)筛选出HCO_(3)^(-),TDS和Mg^(2+)这3个指标作为模型的判别因子,最大化的保留分类信息;运用最小二乘支持向量机算法对20组训练样本进行学习训练,以剩余8组数据作为验证样本,建立基于SR-LSSVM的矿井涌水水源识别模型,并将模型的实测结果与支持向量机、最小二乘支持向量机模型的结果进行对比。结果表明:利用SR-LSSVM模型预测的矿井涌水水源的准确率为100%,显著高于其他模型的预测结果,说明该方法可以对矿井涌水水源进行准确识别;将该模型应用到4个待测样本的识别预测中,判别结果与实际情况完全吻合。研究认为基于SR-LSSVM法的水源识别模型与水化学分析法相比能够有效排除干扰因素的影响,精确识别矿井涌水的类型,该方法为矿井水害防治提供一定的依据和参考。
In order to distinguish the source of mine water gushing quickly and accurately,taking the Ningtiaota coal mine as an example,the water quality test is made of the Sarawusu Formation aquifer,the weathered bedrock aquifer of the Zhiluo Formation,the burnt rock aquifer and the stagnant water in goaf,and the hydrochemical types of different aquifers are analyzed.And the concentration of Na^(+)+K^(+),Ca^(2+),Mg^(2+),Cl^(-),SO_(4)^(2-),HCO_(3)^(-) and TDS is selected as the discriminant index of water source identification.The stepwise regression analysis is used to select HCO_(3)^(-),TDS and Mg^(2+)as the discriminant factors of the model to maximize the retention of classification information.The least square support vector machine algorithm for 20 groups of training samples.The remaining 8 groups of samples are selected to establish a mine water source identification model based on SR-LSSVM,with the measured results compared of the model and of support vector machine and least square support vector machine models.The results show that the accuracy rate of mine water gushing source predicted by the SR-LSSVM model is 100%,which is significantly higher than the prediction results of other models,indicating that this method can accurately identify mine water gushing sources.This model is used to identify and predict four samples to be tested,the discrimination results are completely consistent with the actual situation.It turns out that the water source identification model based on the SR-LSSVM method can effectively eliminate the influence of interference factors and accurately identify the type of mine water inrush,which provides a certain basis and reference for the prevention and control of mine water hazards.
作者
侯恩科
姚星
文强
HOU Enke;YAO Xing;WEN Qiang(College of Geology and Environment,Xi’an University of Science and Technology,Xi’an 710054,China;Xi’an Research Institute,China Coal Technology and Engineering Group,Xi’an 710054,China)
出处
《西安科技大学学报》
CAS
北大核心
2021年第4期624-631,共8页
Journal of Xi’an University of Science and Technology
基金
国家自然科学基金项目(41472234)。
关键词
水源识别
矿井水害
水化学特征
逐步回归
最小二乘支持向量机
water source identifucation
mine water disaster
hydrochemical characteristics
stepwise regression
least squares support vector machine
