摘要
影响公路边坡稳定性的因素众多,各因素与边坡稳定性之间的关系是未知且非线性的,常规简单数学模型难以对边坡稳定性进行有效评价。通过主成分分析法(Principal Component Analysis,PCA)去除影响因素间的相关性,将影响边坡稳定性的6个因素(容重、内聚力、内摩擦角、边坡角、边坡高度、孔隙水压力比)进行主成分提取,得到4个主成分,作为遗传算法(Genetic Algorithms,GA)优化支持向量机(Support Vector Machine,SVM)模型的输入变量,以边坡稳定性作为输出变量,最终建立基于PCA-GA-SVM的公路边坡稳定性评价模型。通过对比检验样本的评价值和实际值,模型的最大绝对误差为0.0921,最大相对误差为9.21%,满足实际工程的要求。
There are many factors that affect the stability of highway slope,and the relationship between each factor and the stability of slope is unknown and nonlinear.So it is difficult to evaluate the stability of slope effectively by adopting simple mathematical models.Firstly,by adopting Principal Component Analysis(PCA),we extracted four principal components from the six slope stability-affecting factors,such as bulk density,cohesion,internal friction angle,slope angle,slope height,and pore water pressure ratio.Secondly,these four principal components are used as input variables of Support Vector Machine(SVM)model optimized by Genetic Algorithm(GA),and the slope stability is taken as the output variable.Finally,the evaluation model of highway slope stability based on PCA-GA-SVM is established.Comparison results between the evaluation and the actual values of the testing samples show that the maximum absolute error of the model is 0.0921 and the maximum relative error is 9.21%,indicating that the model meets the requirements of practical engineering and can provide a basis for highway slope prevention and control.
作者
牛鹏飞
NIU Pengfei(School of Exploration Technology and Engineering,Hebei GEO University,Shijiazhuang 050011,China)
出处
《防灾科技学院学报》
2019年第4期8-14,共7页
Journal of Institute of Disaster Prevention
基金
国家自然科学基金(41807231)
河北省自然科学基金(D201903182)
河北省教育厅青年基金(QN2019196)
河北省教育厅在读研究生创新能力培养资助项目(CXZZSS2019115)
河北地质大学第十六届学生科技基金后补助科研项目(KAD201906)
关键词
公路边坡
主成分分析
遗传算法
支持向量机
稳定性
highway slope
Principal Component Analysis
Genetic Algorithm
Support Vector Machine
stability
