期刊
NATURAL HAZARDS
卷 72, 期 2, 页码 405-422出版社
SPRINGER
DOI: 10.1007/s11069-013-1014-2
关键词
Global stability; Room and pillar coal mines; Retreat mining; Logistic regression; Fuzzy logic; Sugeno inference system
Global stability is a necessary prerequisite for safe retreat mining and one of the crucial and complex problems in room and pillar mining, so its prediction plays an important role in the safety of retreat mining and the reduction of pillar failure risk. In this study, we have tried to develop predictive models for anticipating global stability. For this purpose, two of the most popular techniques, logistic regression analysis and fuzzy logic, were taken into account and a predictive model was constructed based on each. For training and testing of these models, a database including 80 retreat mining case histories from 18 room and pillar coal mines, located in West Virginia State, USA, was used. The models predict global stability based on the major contributing parameters of pillar stability. It was found that both models can be used to predict the global stability, but the comparison of two models, in terms of statistical performance indices, shows that the fuzzy logic model provides better results than the logistic regression. These models can be applied to identify the susceptibility of pillar failure in panels of coal mines, and this may help to reduce the casualties resulting from pillar instability. Finally, the sensitivity analysis was performed on database to determine the most important parameters on global stability. The results revealed that the pillar width is the most important parameter, whereas the depth of cover is the least important one.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据