论文部分内容阅读
针对网络控制系统中存在的时变采样周期与时延,通过矩阵Jordan变换与分解,将采样周期和时延的不确定性转变为系统结构参数的不确定性,建立了离散时间凸多面体不确定系统模型。在此基础上,首先设计观测器,保证系统状态和故障估计收敛于实际值。接着,根据估计的故障,设计了主动动态输出反馈鲁棒容错控制器,给出了执行器发生故障时,系统能保持渐近稳定的充分条件。将控制器设计问题转化为以线性矩阵不等式形式为约束条件的凸优化问题,进而得出了最优H∞鲁棒容错控制器参数的具体表达式。数值仿真验证了提出的设计方法的有效性。
Aiming at the time-varying sampling period and delay in networked control system, the uncertainty of sampling period and delay is transformed into the uncertainty of system structure parameters by matrix Jordan transformation and decomposition, and the discrete-time convex polytope is established indefinitely System model On this basis, the observer is first designed to ensure that the system status and fault estimates converge to the actual values. Then, based on the estimated fault, a robust fault-tolerant controller with active dynamic output feedback is designed, and the sufficient condition for the system to be asymptotically stable is given when the actuator fails. The controller design problem is transformed into a convex optimization problem which is constrained by the linear matrix inequality, and then the concrete expression of the parameters of the optimal H∞ robust fault-tolerant controller is obtained. Numerical simulation verifies the effectiveness of the proposed design method.