摘要
结合支持向量机与遗传算法,提出了一种边坡最危险滑动面识别的新方法。这种方法基于正交设计和极限平衡分析获得学习样本,通过支持向量机学习,从而获得最危险滑动面与安全系数之间的非线性映射关系,再用遗传算法从全局空间上搜索,进行最危险滑动面的识别,并同时获得对应的最小安全系数,该方法计算速度快、效率高。给出了两个算例,结果是令人满意的。
A new approach to recognize critical slip surface is proposed by combining the support vector machine and genetic algorithm.The learning and testing samples produced in orthogonal experiment are used to train the support vector machine. Thusl the support vector machine is used to describe the relationship between slip surface and factor of safety. Then genetic algorithm is adopted to search critical slip surface in their global ranges. This approach was applied to two examples. The results are satisfactory.
出处
《岩土力学》
EI
CAS
CSCD
北大核心
2006年第11期2011-2014,共4页
Rock and Soil Mechanics
基金
中国科学院武汉岩土力学研究所岩土力学重点实验室开放基金课题(No.Z110405)
浙江省高校青年教师资助计划项目
关键词
支持向量机
位移反分析
遗传算法
有限元
slope
critical slip surface
support vector machine
genetic algorithm
limit equilibrium analysis
