线性大系统基于脉冲型信号的加权闭环迭代学习控制

Pulse Signal-Based Weighting Closed-Loop Iterative Learning Control for Large-Scale Linear Systems

摘要: 本文针对线性大工业过程的稳态递阶优化控制，当系统存在初始状态漂移时，研究了基于脉冲型信号的加权闭环PD-型迭代学习控制算法，并在Lebesgue-p范数意义下利用推广的Young卷积不等式分析了算法的收敛性。数字仿真表明，引入脉冲型信号的加权PD-型迭代学习控制算法能有效地减小初始状态漂移引起的跟踪误差，并能显著改善系统暂态响应的动态品质，如抑制超调，加快响应速度，缩短过渡时间等，有效的验证了理论分析的正确性。

Abstract: In this paper, the pulse signal-based closed-loop PD-type iterative learning control algorithms are proposed for steady-state hierarchical optimizing control of large-scale linear industrial processes. The convergence of the updat- ing rules is analyzed in the sense of Lebesgue-p norm by using the generalized Young inequality of convolution integral. Numerical simulations show that the PD-type iterative learning control algorithms presented by a pulse signal may ef- fectively suppress the tracking error caused by the initial state shifts and simultaneously can significantly improve the transient performance of the system such as with no or less overshooting, quick transient response, short setting time and so on. Furthermore, it exhibits the validity of the theoretical analysis.