| 基于连续型Hopfield神经网络的系统辨识 |
| |
| 引用本文： | 李风军,陆桂琴.基于连续型Hopfield神经网络的系统辨识[J].宁夏师范学院学报,2014(6):63-65. |
| |
| 作者姓名： | 李风军 陆桂琴 |
| |
| 作者单位： | 宁夏大学数学与计算机学院,宁夏银川750021 |
| |
| 基金项目： | 国家自然科学基金项目（61063020,11261042）. |
| |
| 摘 要： | 连续型Hopfield神经网络(CHNN)可用于优化计算,但其会遭遇较复杂的参数辨识问题.为了较好地解决这一问题,将擅长全局搜索的蚁群-粒子群混合算法用于对系统参数的最优化选取.再将此混合算法与CHNN有机结合,更好地解决参数辨识问题,且能有效避免CHNN在应用过程中陷入局部最优解.最后,将理论结果应用于求解TSP问题来验证其有效性.
|
| 关 键 词： | CHNN神经网络 系统辨识 蚁群-粒子群混合算法 |
System Identification Based on Continuous Hopfield Neural Network |
| |
| LI Fengjun,LU Guiqin.System Identification Based on Continuous Hopfield Neural Network[J].Journal of Ningxia Teachers College,2014(6):63-65. |
| |
| Authors: | LI Fengjun LU Guiqin |
| |
| Institution: | ( School of Mathematics and Computer Science ,Ningxia University, Yinchuan, Ningxia ,750021 ) |
| |
| Abstract: | Continuous Hopfield Neural Network （CHNN） can be used for optimization calculation, but it will encounter the problem that the parameter identification is more complicated. In order to solve the problem, we use the ant algorithm with particle swarm algorithm to optimize the system parameters ,which are good at global search. Furthermore, combine the hybrid algorithm with the CHNN network, it can solve the parameter identification problem, and can effectively prevent falling into local optimal solution with the application of CHNN. Finally, the theoretical results are applied to solve the TSP problem to verify its effectiveness. |
| |
| Keywords: | Continuous hopfield neural network System identification Ant algorithm and particle swarm hybrid algorithm |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
