基于高斯过程分类的堤基管涌发生可能性识别方法  

Identification method of levee foundation piping possibility based on Gaussian process classification

作　　者：肖义龙 林利森 赵鹏 

机构地区：[1]江西省水投华东设计有限公司,江西南昌300029 [2]江西省安澜工程咨询有限公司,江西南昌330001 [3]西安中粮工程研究设计院有限公司,陕西西安710082

出　　处：《中国水能及电气化》2016年第9期55-58,22,共5页China Water Power & Electrification

摘　　要：管涌是导致堤基渗透变形破坏的主要原因之一,其影响因素众多且各因素之间又存在高度的非线性关系。因此,本文提出了一种基于高斯分类模型的堤防管涌发生可能性识别的新方法。首先,通过数学实例对比展示该方法的优越性,进而将该方法应用于工程实例中,结果表明该方法是可行的,且模型具有参数自适应确定、容易实现且识别精度高等优点,对于迅速识别堤基管涌发生的可能性具有重要的实际应用价值。Piping is one of main causes of levee foundation seepage deformation and destruction. It is affected by many factors, and there is a highly nonlinear relationship among all factors. Therefore, a new method of recognizing levee piping possibility based on Gaussian disaggregated model is proposed. Firstly, the superiority of the method is displayed through math example contrast, thereby the method is applied in engineering example. The result shows that the method is feasible, the model has advantages of parameter adaptive determination, easy realization, high recognition precision, etc. It has important practical application value to quickly recognize the possibility of levee foundation piping.

关 键 词：高斯过程分类 堤防工程 管涌识别 机器学习 

分 类 号：TV871[水利工程—水利水电工程]