期刊
出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/j.ieri.2012.06.021
关键词
dam; piping; Distance Discriminant Analysis Model; SVM
In the paper, mechanism for generation of piping in dam and key factors that affect the generation of piping are analyzed; eight measured indexes are selected as basis of prediction; such prediction methods as Distance Discriminant Analysis Model and SVM (Support Vector Machine) are established for piping in dam, meanwhile, contrastive analysis for the method has been carried out to neural network method. According to the study of twenty-three actual cases of piping projects in dam, it is showed that Distance Discriminant Method and SVM prediction model are with good performance. SVM that is based on neural network kernel function and Radial Basis Kernel Function has much higher prediction accuracy and SVM method is one effective method to solve issue of piping prediction in dam, which can be used in actual projects. (C) 2012 Published by Elsevier B.V. Selection and peer review under responsibility of Information Engineering Research Institute
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据