The application of support vector machine in geotechnical engineering

2018 International Conference on Civil and Hydraulic Engineering

土木建筑工程;水利工程

Ma, Guotao^1,2,3 ; Chao, Zhiming^1 ; Zhang, Ye^2 ; Zhu, Yanjie^2 ; Hu, Hengyang^4

Chengdu Derek Super-computational Technology Co. Ltd, Chengdu, Sichuan

610000, China^1

School of Engineering, University of Warwick, Coventry

CV4 7AL, United Kingdom^2

Faculty of Geoscience and Environmental Engineering, Southwest Jiaotong University, Chengdu, Sichuan

610031, China^3

Southwest Subsidiary Company of China Airport Construction Group Corporation, Chengdu, Sichuan

610202, China^4

关键词: Analytical method;    Basic principles;    Data collection;    Deformation displacements;    Earthquake mitigations;    Machine learning techniques;    Physical strength;    Storage abilities;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/189/2/022055/pdf
DOI  :  10.1088/1755-1315/189/2/022055

学科分类：土木及结构工程学

来源: IOP

【 摘 要 】

With the improvement of data collection and storage ability, numerous data are accumulated in the field of geotechnical engineering, which provides the opportunity for the application of the machine learning techniques. An increasing number of researchers adopted machine learning technique to solve the problems which cannot be addressed by using traditional methods. The most advanced machine learning algorithm, Support Vector Machine (SVM), has been widely utilized in geotechnical engineering. The study aims to review the analytical method and application of SVM in geotechnical engineering. Firstly, the basic principles of SVM are introduced. Secondly, the application of the SVM algorithms is presented. The review suggests that SVM can be effectively used for classifying rock and soil mass, predicting the slopes stability accurately and deformation displacement. Meanwhile, physical strength parameters and the models used for earthquake mitigation that are produced by using SVM are the closest to real value.