基于CM-GCT的尾矿坝在线监测位移分级预警阈值研究  

Research on Graded Early-warning Threshold of Tailings DamOnline Monitoring Displacement Based on CM-GCT

作　　者：袁友翠 胡少华[1] 吴浩 昝军 杨文东[1] 周琳 YUAN Youcui;HU Shaohua;WU Hao;ZAN Jun;YANG Wendong;ZHOU Lin(School of Safety Science and Emergency Management,Wuhan University of Technology,Wuhan 430070,China;不详)

机构地区：[1]武汉理工大学安全科学与应急管理学院,湖北武汉430070 [2]华中师范大学城市与环境科学学院,湖北武汉430079 [3]湖北省安全生产应急救援中心,湖北武汉430070

出　　处：《武汉理工大学学报（信息与管理工程版）》2022年第5期758-765,共8页Journal of Wuhan University of Technology：Information & Management Engineering

基　　金：国家大坝安全工程技术研究中心开放基金项目(CX2019B014);国家自然科学基金项目(51979208);国家“十三五”重点研发计划重点专项项目(2017YFC0804600).

摘　　要：为准确监控尾矿坝运行过程中的异常状态,确定大坝整体变形在不同危险程度下位移的分级预警阈值,基于尾矿坝位移在线监测时间序列,通过云模型(CM)获得表征单测点水平和垂直位移定性概念的特征值,根据云综合算法将单测点的位移特征值进行概念提升,获得表征尾矿坝整体水平和垂直位移定性概念的泛概念树(GCT),依据“3 En”原则计算位移的正常运行值,确定基于CM-GCT的尾矿坝位移分级预警阈值,利用某尾矿坝表面位移监测数据进行实例验证。研究结果表明:该尾矿坝在水平方向位移的分级预警阈值分别为:4.75 mm/d、7.3 mm/d、10.95 mm/d,在垂直于坝轴方向位移的分级预警阈值分别为:11.44mm/d、17.6 mm/d、26.4 mm/d;进一步与单测点位移的分级预警阈值和基于熵-云耦合模型确定的分级预警阈值进行对比,结果表明本文确定的综合分级预警阈值更精确,安全裕度更高。In order to accurately monitor the abnormal state of the tailings dam during operation,the graded early warning thresholds of dam deformation under different dangerous degrees are determined,based on the online monitoring time sequence of each monitoring point,a cloud model(CM)is used to obtain the characteristic values representing the horizontal and the vertical displacement.The values are conceptually improved according to the cloud comprehensive algorithm,and the general concept trees(GCT)of the characteristic values of horizontal and vertical displacement of the whole tailings dam are obtained,then the normal operation value of the displacement is determined according to the"3 En principle",therefore,the graded early-warning threshold model based on CM-GCT is constructed,meanwhile,an online monitoring data of the surface displacement of a tailings dam are used to verify the accuracy and rationality of graded early-warning threshold.The research results show that the four-level early-warning thresholds for the horizontal displacement are:4.75 mm/d,7.3 mm/d,10.95 mm/d,and the graded early-warning thresholds for the vertical displacement are:11.44 mm/d,17.6 mm/d,26.4 mm/d.Furthermore,the graded early-warning threshold calculated based on CM-GCT in this paper is compared with the graded early-warning threshold based on the entropy-cloud coupling model and determined by single monitoring point displacement,the results show that the graded early-warning threshold calculated in this paper is more accurate and reasonable.

关 键 词：尾矿库 分级预警阈值 云模型 泛概念树 特征值 

分 类 号：X936[环境科学与工程—安全科学]