爆破振动对边坡稳定性影响的FA-IGA-LSSVM模型

FA-IGA-LSSVM model of the influence of blasting vibration on slope stability

为对矿山开采爆破过程中边坡的稳定性进行预测,将因子分析、免疫算法及最小二乘支持向量机相结合,共提取爆破振幅、主频率、主频率持续时间、岩石重度、粘聚力、边坡角、边坡高度7个影响指标.通过因子分析对样本数据进行降维,提取出一个公共因子.利用实际测量的29组样本数据对模型进行训练,构建基于因子分析和IGA-LSSVM的边坡稳定性预测模型;采用回代估计法对模型进行检验,误判率为3/29.使用其他5组样本检验模型的泛化能力,同时与基本最小二乘支持向量机进行对比,结果表明:所得模型的预测精度高于基本最小二乘支持向量机,预测结果的误判率为0.

To predict the stability of the slope in the process of mining blasting, this paper made combination of the factor analysis, immune algorithm and least squares support vector machine（SVM）, and selected seven factor index as the influencing factors, which includes blasting amplitude, main frequency, the main frequency duration, heavy rock, cohesive force, internal friction angle, slope angle, and slope height. To extract a common factor, factor analysis was carried out on the sample data dimension reduction. The model was trained by the 29 sample data and the factor analysis and IGA-LSSVM were used to construct the slope stability forecast model;The model was tested by using the method of return generation, and the miscarriage rate was 3/29. The generalization ability of the model is tested using the other 5 sets of samples, at the same time, and it was compared with SVM. The results show that the accuracy of the model is higher than that of SVM, and the false positive rate of the prediction is 0.