摘要
针对山区地下开采自然边坡稳定性分析问题,根据大量工程实测资料,建立了人工神经网络预测模型。通过对人工神经网络算法的改进,选取适当的动量项系数及变步长方法,对已有的实测资料进行了训练和测试,并对丁家河磷矿自然边坡稳定性进行了具体的预测分析,理论计算结果与工程实际情况一致。分析结果表明,所建立的理论模型可用于山区磷矿开采自然边坡稳定性预测分析。
Based on the results of the statistical analysis of a large amount of measured dada in slope engineering, a theoretical model for natural slope failure for underground mining of phosphorus ore-deposit in mountainous areas is established by using the artificial neural networks theory and applied to predict the stability factor of natural slope in Dingjiahe phosphorus mine. The agreement of the theoretical results with the actual situation of natural slope shows that the proposed model is satisfactory; and the theoretical models obtained are valid; and thus can be effectively used for predicting the slope failure due to underground mining of phosphorus ore-deposit in mountainous areas.
出处
《岩土力学》
EI
CAS
CSCD
北大核心
2006年第9期1563-1566,共4页
Rock and Soil Mechanics
基金
河北省科技攻关计划项目(No.3213810)
中国科学院岩土力学重点实验室资助课题(No.Z110406)
河北省教育厅科研项目(No.2004308)。
关键词
采矿工程
岩石力学
磷矿
人工神经网络
自然坡
mining engineering
rock mechanics
phosphorus ore-deposit
artificial neural networks
natural slope
