摘要
为准确可靠地预测岩爆灾害,将随机森林(RF)与层次分析法(AHP)结合,在RF分析指标重要性的基础上优化AHP法,构建RF-AHP指标权重计算方法;结合云模型理论,建立基于RFAHP-云模型的岩爆烈度分级预测模型;通过文献调研法,建立包含301组岩爆工程实例的数据库作为岩爆烈度分级预测的样本数据,并分析25组预测样本的岩爆预测结果。结果表明:所提模型预测准确率达88%以上,可判定预测样本的岩爆烈度等级;经验证,作为预测模型核心的RF-AHP指标权重计算方法具备一定的合理性。
In order to accurately and reliably predict rockburst disasters,AHP was optimized based on importance of analysis index of RF,and an RF-AHP weight calculation method was constructed.Then,a prediction model of rockburst intensity classification based on RF-AHP-Cloud model was established.Finally,through literature survey,a database containing 301 groups of engineering instances was established as sample data for rockburst prediction,and prediction results of 25 sets of samples were analyzed.The results show that the proposed model has an prediction accuracy of more than 88%,and it can determine rockburst intensity grade of samples.And rationality of RF-AHP index weight calculation method as core of prediction model is also verified.
作者
田睿
孟海东
陈世江
王创业
石磊
TIAN Rui;MENG Haidong;CHEN Shijiang;WANG Chuangye;SHI Lei(Institute of Mining Engineering,Inner Mongolia University of Science and Technology,Baotou Inner Mongolia 014010,China;Inner Mongolia Institute of Geological Environmental Monitoring,Hohhot Inner Mongolia 010020,China)
出处
《中国安全科学学报》
CAS
CSCD
北大核心
2020年第7期166-172,共7页
China Safety Science Journal
基金
国家自然科学基金资助(51564038,51464036)
内蒙古自治区自然科学基金资助(2018MS05037)
内蒙古自治区博士研究生科研创新项目(B20171012702)。
