摘要
工作面涌水量的准确预测是预防煤层顶板水害的关键。通过调查收集凉水井矿区14个工作面的实测涌水量数据,利用多元回归模型进行公式拟合,使用层次分析法(AHP)与熵权法(EWM)相结合的方式确定因素的权重,得到基于多元回归模型与层次分析法、熵权法结合的工作面涌水量预测方法。与其他涌水量预测公式(大井法、水文地质比拟法)相比,基于多元回归模型与层次分析法、熵权法结合的涌水量预测方法预测误差最小,平均误差仅为11.6%。以凉水井煤矿42201、42202、42203、42205工作面为例,进行工作面涌水量预测,取得较好的应用效果,进一步验证了公式的适用性与准确性。该方法可准确预测凉水井矿区工作面涌水量,可作为该矿区涌水量预测的一种新方法。
The accurate prediction of water inflow in working face is the key to prevent water damage of coal seam roof.Through the investigation and collection of the measured water inflow data of 14 working faces in Liangshuijing mining area,the formula fitting is carried out by using the multiple regression model,and the weight of factors is determined by using the combination of analytic hierarchy process(AHP)and entropy weight method(EWM).The prediction method of water inflow of working faces based on the combination of multiple regression model,analytic hierarchy process and entropy weight method is obtained.Compared with other water inflow prediction formulas(big well method and hydrogeological analogy method),the prediction error of the water inflow prediction method based on the combination of multiple regression model,analytic hierarchy process and entropy weight method is the smallest,and the average error is only 11.6%.Taking Liangshuijing coal mine 42201,42202,42203 and 42205 working faces as examples,the prediction of water inflow of working faces has achieved good application results,and further verified the applicability and accuracy of the formula.This method can accurately predict the water inflow of the working face in Liangshuijing mining area,and can be used as a new method to predict the water inflow in this mining area.
作者
王档良
房亚飞
邓国伟
高成跃
司湘
WANG Dang-liang;FANG Ya-fei;DENG Guo-wei;GAO Cheng-yue;SI Xiang(School of Resources and Earth Sciences,China University of Mining and Technology,Xuzhou221116,China;Shaanxi Energy Liangshuijing Mining Co.,Ltd.,Shenmu 719319,China)
出处
《煤炭科技》
2022年第4期85-92,共8页
Coal Science & Technology Magazine
基金
国家自然科学基金项目(41877238)。
关键词
改进多元回归模型
涌水量预测
预防煤层顶板水害
improved multiple regression model
prediction of water inflow
prevention of water damage to coal seam roof
