本文關鍵詞： 隨機森林 邊坡穩(wěn)定性 離子型稀土 預測模型　出處：《礦業(yè)研究與開發(fā)》2017年04期 　論文類型：期刊論文

【摘要】：為對離子型稀土原地浸礦邊坡進行穩(wěn)定性預測,結合贛南離子型稀土礦山42個邊坡實例,選取重度、黏聚力、內(nèi)摩擦角、邊坡角、孔隙壓力比5個影響因子作為輸入,邊坡狀態(tài)作為輸出,通過隨機森林算法建立邊坡穩(wěn)定性影響因素與邊坡穩(wěn)定狀態(tài)之間的非線性關系。利用30組邊坡穩(wěn)定性數(shù)據(jù)作為隨機森林預測模型的訓練數(shù)據(jù)集,進行模型的學習訓練;用另12組邊坡穩(wěn)定性數(shù)據(jù)作為預測模型的測試數(shù)據(jù),通過訓練好的邊坡穩(wěn)定性預測模型進行測試。結果表明,隨機森林預測模型精度高,能夠為離子型稀土原地浸礦邊坡的災害防治工作提供指導。
[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.
【作者單位】： 江西理工大學建筑與測繪工程學院;
【分類號】：TD865;TD854.6
【正文快照】： 離子型稀土原地浸礦具有保護地表植被、不占用地表空間、較高的資源回收利用率等優(yōu)點,被國家強制要求使用[1-2]。在浸礦過程中,采場滑坡事故時有發(fā)生,對礦區(qū)的經(jīng)濟和環(huán)境造成了嚴重的損失[3]。離子型稀土原地浸礦邊坡穩(wěn)定性受多種因素的影響,其中重要的因素有重度、黏聚力、內(nèi)

【相似文獻】

相關期刊論文 前1條

1 溫廷新;張波;邵良杉;;煤與瓦斯突出預測的隨機森林模型[J];計算機工程與應用;2014年10期

，

本文編號：1485882