| 基于Hopfield神经网络的非线性系统故障估计方法 |
| |
| 引用本文： | 王占山,张恩林,张化光,冯健.基于Hopfield神经网络的非线性系统故障估计方法[J].南京航空航天大学学报,2011(Z1). |
| |
| 作者姓名： | 王占山 张恩林 张化光 冯健 |
| |
| 作者单位： | 东北大学信息科学与工程学院; |
| |
| 基金项目： | 国家自然科学基金重点(61074073,61034005)资助项目; 教育部博士点新教师基金(200801451096)资助项目; 高校基本科研基金(N090404017,N100104102)资助项目 |
| |
| 摘 要： | 针对一类故障参数是线性可分的非线性系统,在参数故障情况下提出了一种基于Hopfield递归神经网络的故障估计方法。利用Hopfield递归神经网络的自学习能力和稳定性理论,将故障参数的估计问题转化为Hop-field神经网络的稳定性问题,克服了现有数值方法存在量化误差和算法收敛性等问题。与自适应观测器和等价空间方法等相关故障参数估计方法相比,具有设计简单、易于实现和适用性宽等特点。仿真结果表明,对于常值故障和时变故障,诊断系统均具有较好的估计效果和动态性能。
|
| 关 键 词： | Hopfield递归神经网络 非线性系统 故障估计 稳定性 |
Fault Estimation Approach for a Class of Nonlinear Systems Based on Hopfield Neural Network |
| |
| Wang Zhanshan,Zhang Enlin,Zhang Huaguang,Feng Jian.Fault Estimation Approach for a Class of Nonlinear Systems Based on Hopfield Neural Network[J].Journal of Nanjing University of Aeronautics & Astronautics,2011(Z1). |
| |
| Authors: | Wang Zhanshan Zhang Enlin Zhang Huaguang Feng Jian |
| |
| Institution: | Wang Zhanshan,Zhang Enlin,Zhang Huaguang,Feng Jian(School of Information Science and Engineering,Northeastern University,Shenyang,110004,China) |
| |
| Abstract: | For a class of nonlinear systems with fault parameters varying linearly with the system states and outputs,fault estimation approach based on Hopfield recurrent neural networks is proposed.Sufficiently using the learning capability and stability theory of recurrent neural networks,the parameter fault estimation problem is converted to the stability problem of the recurrent neural networks.The proposed approach overcomes the shortcomings of quantitative error and convergence of numerical algorithms.Comparing... |
| |
| Keywords: | Hopfield recurrent neural networks nonlinear systems fault estimation stability |
| 本文献已被 CNKI 等数据库收录！ |
