飞行器网络控制系统在线故障检测算法

Online fault detection of networked control systems for flight vehicles

针对飞行器网络控制系统具有网络诱导时延且采样周期较短的情况,提出了一种飞行器网络控制系统在线故障检测算法.考虑未知时延对网络控制系统的影响,建立飞行器网络化控制系统故障检测模型.基于此模型,构造了残差生成器,将故障检测问题转化为最小化优化问题.以矩阵Moore-Penrose逆的形式给出该最小化优化问题的解,在每个采样周期内进行飞行器网络控制系统在线故障检测.数值算例表明:当飞行器网络控制系统存在未知时延时,该故障检测算法对故障敏感,同时对未知时延引起的扰动具有鲁棒性.

An online fault detection algorithm for the networked control system of flight vehicle was pro- posed considering the network-induced delay and the short sampling period. The fault detection model was es- tablished for the impact of the unknown time-delay. Based on the model, the residual generator was construc- ted and the fault detection of networked control systems was converted to the minimum optimization problem. Furthermore, the solution of the problem was given in the form of Moore-Penrose Inverse of matrix, and the fault detection algorithm was optimized in every sampling period. The numerical example shows that the pro- posed fault detection method is sensitive to the fault, and it is also robust to the unknown disturbance caused by the unknown delay.