| 基于RBF神经网络的曲面数据修补研究 |
| |
| 引用本文： | 杨科,廖俊必.基于RBF神经网络的曲面数据修补研究[J].机械,2007,34(1):11-13,23. |
| |
| 作者姓名： | 杨科 廖俊必 |
| |
| 作者单位： | 四川大学,制造科学与工程学院,四川,成都,610065;四川大学,制造科学与工程学院,四川,成都,610065 |
| |
| 摘 要： | 径向基函数(RBF)神经网络具有良好的泛函逼近能力,主要探讨了将RBF神经网络应用于残缺曲面数据修补问题,通过建立适于数据修补的网络模型与采用levenberg-marquardt算法的改进型BP神经网络进行性能比较.结果表明:RBF在进行残缺数据修补时网络收敛速度快于BP神经网络,且修补精度高,适宜于曲面残缺数据的修补.
|
| 关 键 词： | RBF神经网络 数据修补 逆向工程 |
| 文章编号： | 1006-0316（2007）01-0011-04 |
| 修稿时间： | 2006-08-12 |
Research on the data repairing in surface by rbf neural network |
| |
| YANG Ke,LIAO Jun-bi.Research on the data repairing in surface by rbf neural network[J].Machinery,2007,34(1):11-13,23. |
| |
| Authors: | YANG Ke LIAO Jun-bi |
| |
| Affiliation: | Sichuan University, Chengdu 610065, China |
| |
| Abstract: | This paper discusses an application for data repairing in surface by radial basis function(rbf) neural network due to its upstanding appropriate ability.An adaptive model for data repairing is constructed and a performance comparison with improved bp neural network which adopts the levenberg-marquardt(l-m) algorithm is drawn.the results show that the convergence speed of rbf for data repairing is superior to that of bp network and the precision of rbf is high.This approach is suitable for data repairing in surface. |
| |
| Keywords: | rbf neural network data repairing reverse engineering |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
