摘要
在迭代学习控制理论的收敛性分析中,常见的初始条件是迭代初值与期望初值一致,或者迭代初值固定,给出了一类含控制时滞非线性时变系统在任意初值条件下采用开环PD型迭代学习控制算法时的收敛条件。迭代学习采用控制输入与初值同时学习的算法,其中控制输入利用了给定超前法,该算法解决了控制时滞和初值问题。运用算子理论证明了收敛条件,给出了间歇非线性控制时滞过程仿真实例,研究结果说明了该算法的有效性。
In the iterative learning control theory, the eommon initial condition is that the iterative initial value is fixed or consistent with the desired initial value. The convergence condition of the open-loop PDtype iterative learning control for a class of nonlinear time-varying systems with control delay and arbitrary initial value is studied. The control input and the initial value learning simultaneously is used in the iterarive learning, and the control input signal is given in advance. The algorithm solves the problem of control delay and initial value. The convergence of the learning algorithm is proved by using the operator theory. A simulation example of batch nonlinear control delay processes is given also. The research results illustrate the effectiveness of the algorithm.
出处
《兵工学报》
EI
CAS
CSCD
北大核心
2010年第6期837-841,共5页
Acta Armamentarii
基金
教育部博士点基金(20070288022)
江苏省自然科学基金(BK2008404)
关键词
自动控制技术
ILC
初始状态
收敛条件
算子理论
automatic control technology
ILC
initial state
convergence condition
operator theory