LS-SVM在CFG桩复合地基承载力预测中的应用

Application of LS-SVM to prediction of bearing capacity of cement-flyash-gravel pile composite foundation

分析了CFG桩复合地基承载力的主要影响因素,提出了CFG桩复合地基承载力预测的一种新方法,即最小二乘支持向量机方法。根据有限的学习样本,建立了CFG桩复合地基承载力与其影响因素之间的非线性关系。结果表明:最小二乘支持向量机具有优秀的小样本数据学习能力和预测能力,将其用于CFG桩及其它刚性桩复合地基承载力的预测是可行的。

There are a lot of factors that influence the bearing capacity of composite foundation,and the relationship between them is complex and nonlinear.Based on the study of main factors that have great influence on bearing capacity of cement-flyash-gravel（CFG） pile composite foundation,the least squares support vector machine（LS-SVM） model of bearing capacity of composite foundation was established.The results show that the model has excellent learning ability and generalization and can provide accurate data prediction only with fewer observed samples.It is proved that the new method is a promising method for the determination of bearing capacity of CFG pile and other rigid piles composite foundation.