摘要
针对离散线性系统,研究批次长度随机变化的反馈辅助PD型量化迭代学习控制问题.考虑系统信号经量化后传输到控制器或执行器的情况,给出两种量化方案:跟踪误差信号量化和控制输入信号量化.基于两种不同的量化信号,在批次长度和初始条件随机变化前提下设计反馈辅助PD型迭代学习控制算法.采用扇形界的处理方法和堆积系统框架,推导数学期望下的学习收敛条件:在误差信号量化情况下,所提出控制算法可以保证跟踪误差渐近收敛到零;在控制输入信号量化情况下,所提出控制算法能够保证跟踪误差有界收敛.仿真示例对比验证了两种量化方案下所提出方法的有效性和优越性.
The feedback-assisted PD-type quantized iterative learning control problem is studied for discrete linear systems with iteration-varying trial lengths. Considering that the system signal is transmitted to the controller or actuator after being quantized. Two quantization schemes are given, including tracking error signal quantization and control input signal quantization. In the case of iteration-varying trial lengths and iteration-varying initial state conditions, a feedback-assisted PD-type update law is developed based on the quantized signal. The learning convergence condition under mathematical expectations derived with the sector bound method and the lifting representation:tracking error signal quantization can obtain zero tracking error, and control input signal quantization only guarantee that the tracking error converges to a bound. Simulation examples are provided to demonstrate the effectiveness and superiority of the proposed scheme under the two quantization schemes.
作者
王晶
周楠
王森
沈栋
李伯群
WANG Jing;ZHOU Nan;WANG Sen;SHEN Dong;LI Bo-qun(College of Information Science and Technology,Beijing University of Chemical Technology,Beijing 100029,China;School of Mathematics,Renmin University of China,Beijing 100872,China;School of Electronic and Information Engineering,University of Science and Technology LiaoNing,Anshan 114051,China)
出处
《控制与决策》
EI
CSCD
北大核心
2021年第10期2569-2576,共8页
Control and Decision
基金
国家自然科学基金项目(61973023,61573050,61673045)
北京市自然科学基金项目(4202052)。
关键词
迭代学习控制
反馈辅助PD策略
数据量化
随机变批次长度
初始状态条件
iterative learning control
feedback-assisted PD strategy
data quantization
iteration-varying lengths
initial state conditions
作者简介
通讯作者:王晶(1972−),女,教授,博士生导师,从事数据驱动的复杂过程建模、优化、控制及故障诊断等研究,E-mail:jwang@mail.buct.edu.cn;周楠(1994−),女,硕士生,从事鲁棒迭代学习控制的研究,E-mail:zhounan_zara@163.com;王森(1994−),男,博士生,从事鲁棒控制、迭代学习控制、网络系统控制等研究,E-mail:wangsmmm@126.com;沈栋(1982−),男,教授,博士,从事迭代学习控制、复杂通信条件的网络控制系统等研究,E-mail:dshen@ieee.org;李伯群(1970−),男,教授,博士,从事工业过程建模、智能解耦控制等研究,E-mail:lbqhylyxab@163.com.
