不确定时滞系统的PD型迭代学习控制算法  

PD-type learning algorithm for uncertain control time-delay systems

作　　者：张严心[1] 徐健洲[1] 

机构地区：[1]北京交通大学电子信息工程学院,北京100044

出　　处：《北京交通大学学报》2012年第2期70-73,84,共5页JOURNAL OF BEIJING JIAOTONG UNIVERSITY

基　　金：国家自然科学基金资助项目(60904014)

摘　　要：针对不确定时滞系统,在网络时滞范围已知情况下,采用改进PD型迭代学习控制算法补偿网络时滞.在初态是严格重复时,给出这类系统的极限轨迹和迭代输出收敛于该极限轨迹的充分条件.并与P型迭代学习控制算法进行比较.仿真结果表明改进后的PD型迭代学习控制算法能够有效地补偿此类时滞.当网络时滞范围变窄时,能够更加精确跟踪极限轨迹.在相同迭代次数情况下,PD型迭代学习控制算法比P型迭代学习控制算法能更快收敛于极限轨迹.In the paper,for a class of NCS with uncertain time delay,a PD-type iterative learning algorithm（ILC） is studied to compensate it.Based on the strict repetition of the initial state,the sufficient conditions which guarantee the uniform convergence of the ILC are given.And the limit output trajectories generated by the action of the ILC are also presented.Then,comparing with the efficiency of the P-type ILC algorithm,it is shown that the PD-type ILC is more effective to compensate the time delay.For the case that the range of the time delay becomes smaller,it can track the output trajectories precisely than the P-type ILC algorithm.Moreover,under the same number of iteration,the PD-type ILC algorithm can track the state trajectories faster than the PD-type ones.

关 键 词：网络时滞 迭代学习控制 严格重复 不确定时滞系统 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]