基于输出时滞方法的非均匀采样数据系统传感器故障检测(英文)  

An output delay approach to sensor fault detection fornon-uniformly sampled-data systems

作　　者：邱爱兵[1,2] 姜斌[2] 

机构地区：[1]南通大学电气工程学院,江苏南通226019 [2]南京航空航天大学自动化学院,江苏南京210016

出　　处：《控制理论与应用》2010年第12期1757-1765,共9页Control Theory & Applications

基　　金：Supportedby Key Program of National Nature Science Foundation of China(60934009)

摘　　要：研究一般非均匀采样数据系统鲁棒传感器故障检测设计问题.首先,基于输出时滞方法将非均匀采样数据系统转换成具有时变时滞输出的连续系统;然后,选择故障检测滤波器作为残差产生器,并将故障检测设计问题描述成一个多目标优化问题,即连续时间过程噪声和离散时间测量噪声对残差信号的H∞范数小于一个给定值,同时传感器故障对残差信号的l2增益大于一个给定值,基于输入输出方法以矩阵不等式的形式给出该多目标优化问题有解的充分条件;进一步的,提出一个迭代算法来权衡噪声鲁棒性与故障灵敏度,并将矩阵不等式转换成可解的线性矩阵不等式.最后,对某型飞控系统的仿真实验验证了所提方法的有效性.In this paper, the problem of sensor fault detection in general non-uniformly sampled-data systems is inves-tigated. First, the output delay approach is used to model non-uniformly sampled-data systems as continuous ones withtime-varying delay output. Then, based on the input-output approach, a fault detection filter is designed to guarantee notonly the effects of continuous-time process noise and discrete-time measurement noise on residual to satisfy a prescribedH-infinity performance, but also the l-two gain from sensor fault to residual to be greater than a prescribed value. Theexistence condition of such a fault detection filter is given in terms of matrix inequalities（MIs）. Furthermore, an iterativealgorithm which transform MIs into the solvable linear MIs（LMIs） is proposed to make a tradeoff between the noise ro-bustness and fault sensitivity. Finally, simulation results of a certain type of flight control systems are presented to show thevalidity of the proposed method.

关 键 词：非均匀采样数据系统 故障检测滤波器 输出时滞方法 迭代算法 传感器故障 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]