| 混合递阶遗传算法优化小波神经网络的研究 |
| |
| 引用本文： | 冯冀宁,刁哲军,杨晓波,刘红运. 混合递阶遗传算法优化小波神经网络的研究[J]. 系统工程与电子技术, 2005, 27(6): 1112-1114 |
| |
| 作者姓名： | 冯冀宁 刁哲军 杨晓波 刘红运 |
| |
| 作者单位： | 1. 河北师范大学电子工程系,河北,石家庄,050031
2. 石家庄职业技术学院,河北,石家庄,050081 |
| |
| 摘 要： | 在研究小波神经网络结构及其学习算法的基础上,提出了一种以混合递阶遗传算法优化小波神经网络的新方法,混合递阶遗传算法是递阶遗传算法和多元线性回归的结合。此方法可同时优化小波神经网络的结构及其参数,具有学习速度快,精度高的特点。并与传统的BP算法进行了对比,证明了本算法的有效性。
|
| 关 键 词： | 波神经网络 混合递阶遗传算法 优化 |
| 文章编号： | 1001-506X(2005)06-1112-03 |
| 修稿时间： | 2004-04-07 |
Wavelet neural network optimization based on hybrid hierarchy genetic algorithm |
| |
| FENG Ji-ning,DIAO Zhe-jun,YANG Xiao-bo,LIU Hong-yun. Wavelet neural network optimization based on hybrid hierarchy genetic algorithm[J]. System Engineering and Electronics, 2005, 27(6): 1112-1114 |
| |
| Authors: | FENG Ji-ning DIAO Zhe-jun YANG Xiao-bo LIU Hong-yun |
| |
| Abstract: | Based on the wavelet neural network (WNN) training algorithm and geometrical structure, a new WNN optimization algorithm-hybrid hierarchy genetic algorithm (HHGA) is introduced by combining hierarchy genetic algorithm and multivariate linear regression. This algorithm is effective to determine the structure and parameters of WNN based on data and has the merits of high learning speed and higher precision. The method is compared with the traditional BP algorithm. The effectiveness of the algorithm is demonstrated. |
| |
| Keywords: | wavelet neural network hybrid hierarchy genetic algorithm optimization |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
