提出了一种基于支持向量机的露天转地下开采边坡变形模型,有效表达了地下开采扰动引起露天矿边坡变形的非线性变化关系.采用RBF核函数学习现场监测数据,利用交叉验证选择模型参数,通过学习捕捉支持向量,建立模型预测未来变化趋势.将该模型应用于露天转地下开采的杏山铁矿.结果表明,支持向量机对学习样本的拟合精度极高,其预测精度也很高.采用捕捉的支持向量进行预测,便捷快速且有较强泛化能力. A model of slope deformation based on support vector machine (SVM) is proposed, which effectively expresses the nonlinear variation of slope deformation caused by underground mining disturbance.The RBF kernel function is used to study on-site monitoring data and cross-validation model Parameters to learn the support vector to establish a model to predict future trends.This model is applied to the open-pit mining of the Xiaoshan Iron Mine.The results show that the support vector machine to the learning sample fitting accuracy is very high, and its prediction accuracy Very high. Predicted with snapped support vectors, quick and easy to use and fairly generalizable.