随机森林模型在边坡稳定性预测中的应用

发布时间：2018-02-03 00:22

本文关键词： 随机森林边坡稳定性离子型稀土预测模型　出处：《矿业研究与开发》2017年04期 　论文类型：期刊论文

【摘要】：为对离子型稀土原地浸矿边坡进行稳定性预测,结合赣南离子型稀土矿山42个边坡实例,选取重度、黏聚力、内摩擦角、边坡角、孔隙压力比5个影响因子作为输入,边坡状态作为输出,通过随机森林算法建立边坡稳定性影响因素与边坡稳定状态之间的非线性关系。利用30组边坡稳定性数据作为随机森林预测模型的训练数据集,进行模型的学习训练;用另12组边坡稳定性数据作为预测模型的测试数据,通过训练好的边坡稳定性预测模型进行测试。结果表明,随机森林预测模型精度高,能够为离子型稀土原地浸矿边坡的灾害防治工作提供指导。
[Abstract]:In order to predict the stability of ion rare earth in-situ leaching slope, combined with 42 slope examples of ion type rare earth mine in south Jiangxi Province, the heavy, cohesion, internal friction angle and slope angle were selected. The pore pressure ratio of 5 factors is taken as the input and the slope state as the output. The nonlinear relationship between the influencing factors of slope stability and slope stability was established by stochastic forest algorithm, and 30 groups of slope stability data were used as the training data set of stochastic forest prediction model. Carry on the study training of the model; The other 12 groups of slope stability data are used as the test data of the prediction model, and the trained slope stability prediction model is tested. The results show that the stochastic forest prediction model has high accuracy. It can provide guidance for disaster prevention and treatment of ionized rare earth in situ leaching slope.
【作者单位】：江西理工大学建筑与测绘工程学院;
【分类号】：TD865;TD854.6
【正文快照】： 离子型稀土原地浸矿具有保护地表植被、不占用地表空间、较高的资源回收利用率等优点,被国家强制要求使用[1-2]。在浸矿过程中,采场滑坡事故时有发生,对矿区的经济和环境造成了严重的损失[3]。离子型稀土原地浸矿边坡稳定性受多种因素的影响,其中重要的因素有重度、黏聚力、内

【相似文献】