中文摘要：

主要影响角正切是开采沉陷预计的重要参数,对于地表影响范围的确定具有重要的意义。在综合分析主要影响角正切与地质采矿条件之间关系的基础上,运用我国典型的地表移动观测站资料作为学习训练样本,通过合理地选择SVM中的核函数,惩罚参数C,不敏感损失参数ε,建立了主要影响角正切与各影响因素的SVM回归关系模型,并用此模型对检验样本进行预测。结果表明,用支持向量机方法求算主要影响角正切具有较高的精度,可以在实际工程中进行推广应用。该方法可以综合考虑多种因素对主要影响角正切的影响,为今后快速准确地求取主要影响角正切提供了一个新的方法。

英文摘要：

The tangent of the main influencing angle is an important parameter for mining subsidence prediction and surface displacement range demarcation.On the basis of the relationship analysis between the tangent of the main influencing angle and the geologic and mining conditions,the data from the typical ground movement observation stations of China was used as training samples.With a reasonable selection of kernel function of SVM,penalty parameter C and insensitive loss parameter ε,a SVM regression relationship model of the tangent of the main influencing angle and the other factors was established to predict on the samples.The results showed that the SVM prediction method is accuracy and can be applied to the actual engineering.Multiple influence factors could be considered comprehensively in this method,and a new approach to quickly and accurately calculate the tangent of the main influencing angle is proposed for the future.