围岩稳定性评判的支持向量机多分类模型

围岩稳定性评判对指导地下工程的设计和施工具有非常重要的意义.引入支持向量机多分类方法,进行围岩稳定性分类.选用影响围岩稳定性的因素,岩石质量指标、岩石单轴饱和抗压强度、完整性系数、结构面强度系数和地下水渗水量等5项指标作为模型的判别因子,建立了围岩稳定性分类的支持向量机模型.以围岩实测数据作为学习样本进行训练,建立相应判别函数对待判样本进行分类.结果表明,支持向量机模型分类性能良好,预测精度高,是围岩稳定性分类的一种有效方法,可以在实际工程中进行应用.

Multi-Class Support Vector Machine Model on Stability Classification of Surrounding Rocks

The stability classification of surrounding rocks has an important significance for guiding design and construction in under ground engineering.This paper introduces a support vector machine（SVM）method into the stability classification of surrounding rocks.Based on the classification algorithm of support vector machine,the SVM model is established according to main factors with important influence on stability classification,the following factors such as rock quality designation,uneasily compressive strengthen,integrality coefficient,strength coefficient of structural plane and seepage measurement of groundwater are selected as the evaluating indices.Discriminate functions are obtained through training a large set of surrounding rocks samples.The results show that the classification model of SVM has excellent prediction accuracy and can be used in practice.