| 神经网络故障检测算法研究 |
| |
| 引用本文： | 李军,孙金生,王执铨. 神经网络故障检测算法研究[J]. 南京理工大学学报(自然科学版), 2001, 25(6): 606-608,625 |
| |
| 作者姓名： | 李军 孙金生 王执铨 |
| |
| 作者单位： | 南京理工大学自动化系,南京,210094 |
| |
| 摘 要： | 该文采用输出带输入线性项的RBF神经网络来辨识非线性随机系统模型，基于扩展卡尔曼滤波器的设计思想，设计了一个可估计状态滤波协方差阵及状态预测协方差阵上界的次优状态滤波器，以产生残差序列，并可用来进行故障检测。该方法适用于一大类非线性随机系统，具有较大的适用范围。
|
| 关 键 词： | 故障检测 卡尔曼滤波器 神经网络 非线性随机系统 状态滤波器 |
An Algorithm of Fault Detection Using Neural Networks |
| |
| LiJun SunJinsheng WangZhiquan. An Algorithm of Fault Detection Using Neural Networks[J]. Journal of Nanjing University of Science and Technology(Nature Science), 2001, 25(6): 606-608,625 |
| |
| Authors: | LiJun SunJinsheng WangZhiquan |
| |
| Abstract: | The problem of identifying nonlinear stochastic systems using radial basis function (RBF) neural networks is investigated. By utilizing the design idea of extended Kalman filter, a sub optimal state filter is derived.The filter can estimate the upper bounds of the covariance of state estimation and state prediction. Therefore, the residuals for fault detection can be obtained. This method can be applied to a large class of nonlinear stochastic systems. |
| |
| Keywords: | fault detection Kalman filters neural networks nonlinear systems |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
