| 基于前向神经网络的多新息随机梯度辨识算法 |
| |
| 引用本文： | 刘英玉,申东日,陈义俊,李蓉.基于前向神经网络的多新息随机梯度辨识算法[J].哈尔滨商业大学学报(自然科学版),2006,22(2):83-86. |
| |
| 作者姓名： | 刘英玉 申东日 陈义俊 李蓉 |
| |
| 作者单位： | 辽宁石油化工大学,信息工程学院,辽宁,抚顺,113001 |
| |
| 摘 要： | 为了提高动态系统的辨识精度,提出一种基于前馈神经网络的多新息随机梯度辨识算法,它通过动态调整网络权值来提高网络在线辨识性能.由于多新息随机梯度辨识算法利用了系统的当前数据和历史数据,对动态辨识,特别是对具有纯时间延迟动态系统的辨识,较传统的BP算法在辨识精度和收敛速度方面具有更好的效果.仿真结果表明该算法的有效性.
|
| 关 键 词： | 多新息随机梯度辨识算法 前向神经网络 非线性时变系统 |
| 文章编号： | 1672-0946(2006)02-0083-04 |
| 修稿时间： | 2005年10月21 |
Multi-innovation stochastic gradient identification algorithm based on feedforward neural networks |
| |
| LIU Ying-yu,SHEN Dong-ri,CHEN Yi-jun,LI rong.Multi-innovation stochastic gradient identification algorithm based on feedforward neural networks[J].Journal of Harbin University of Commerce :Natural Sciences Edition,2006,22(2):83-86. |
| |
| Authors: | LIU Ying-yu SHEN Dong-ri CHEN Yi-jun LI rong |
| |
| Abstract: | In order to improve the identification accuracy of dynamic system,multi-innovation stochastic gradient identification algorithm based on feedforward neural networks is presented,which can improve the online identification performance of the networks by adjusting its connection weights dynamically.The utilization of thecurrent and past date of the system at the same time makes the presented multiinnovation stochastic gradient identification algorithm more effective than the BP algorithm in view of accuracy and convergence rate.Simulation results showed that the algorithm is effect. |
| |
| Keywords: | multi-innovation stochastic gradient identification algorithm recurrent neural networks nonlinear time varying system |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
