摘要
为解决批次长度随机变化的线性离散时间系统,提出了高阶迭代学习控制算法,实现对期望轨迹的渐进跟踪。将迭代平均算子与PD型迭代学习算法相结合,使控制输入的更新中利用历次迭代信息。通过λ范数分析,证明了所提学习算法可使跟踪误差在数学期望意义下收敛,并且上述算法可放宽对迭代初始状态的要求。实例仿真计算表明,上述算法比现有算法更加有效性,且所提学习算法既可以加快跟踪误差的收敛速度又能减少迭代过程中跟踪误差出现发散的情况。
In order to solve the problem of linear discrete-time system with random batch length,a high-order it⁃erative learning control algorithm is proposed to realize the asymptotic tracking of the desired trajectory.Iterative aver⁃aging operator was combined with PD iterative learning algorithm to make use of iteration information in updating con⁃trol input.Through-λanalysis,the convergence of the tracking error in the mathematical expectation has been proved and the algorithm can relax the requirement for the initial state of each iteration.The simulation example was presen⁃ted to show that the proposed algorithm is more effective than the existing algorithms,and the proposed learning algo⁃rithm can not only accelerate the convergence rate of tracking errors but also reduce the divergence of tracking errors in the iterative process.
作者
岳壮壮
欧阳慧珉
张广明
YUE Zhuang-zhuang;OUYANG Hui-min;ZHUANG Guang-ming(College of Electrical Engineering and Control Science,Nanjing Tech.University,Nanjing Jiangsu 211816,China)
出处
《计算机仿真》
北大核心
2020年第9期239-243,共5页
Computer Simulation
基金
国家自然科学基金(61703202)
江苏省重点研发计划(BE2017164)。
关键词
迭代学习控制
批次长度随机变化
收敛分析
收敛性能
迭代平均算子
Iterative learning control
Randomly varying iteration lengths
Convergence analysis
Convergence performance
Iterative average operator
作者简介
岳壮壮(1994-),男(汉族),安徽省亳州市人,硕士研究生,主要研究领域为迭代学习控制;欧阳慧珉(1982-),男(汉族),江苏省南京市人,副教授,硕士研究生导师,主要研究欠驱动控制;张广明(1965-),男(汉族),江苏人,教授,博士研究生导师,主要研究一体化系统控制。
