测控信号随机时延对迭代学习控制系统的影响

Impact of Random Delays in Control and Measurement Signals on Iterative Learning Control Systems

测控信号通过无线网络传输受到的随机时延严重影响迭代学习控制系统的收敛性能.在一步随机时延模型的基础上,得到包含随机时延影响因子的系统转移矩阵,然后分别针对测控信号发生时延的情况,研究转移矩阵内决定系统收敛速度的特征值,以及决定系统鲁棒收敛性的下三角内其他元素取值的变化情况.由分析可知,测控信号随机时延不仅会降低系统的收敛速度,而且会影响系统的鲁棒收敛性.尤其是控制信号随机时延,它对系统鲁棒收敛性的影响明显强于测量信号随机时延的影响.仿真实验表明了分析结论的正确性.

Random delays of control and measurement signals during transmission over wireless network seriously affect the convergence performance of iterative learning control （ILC） systems. Based on a one-step random delay model, the transition matrix of system is derived, which contains the impact factors of random delays. For different situations of random delays, variation of eigenvalues and other elements in the lower triangular of transition matrix is analyzed, which determine the convergence speed and robust convergence respectively. Analysis shows that the convergence rate is reduced, and the robust convergence is also affected. Especially, the impact of control signal delays on robust convergence is greater than that of measurement signal delays. Simulation results are provided to demonstrate correctness of the conclusion.