摘要
针对一类非线性带扰动系统提出了高阶PID采样迭代学习控制算法,讨论了高阶算法的收敛性问题以及该算法的优势与缺陷。与传统的证明方法不同,利用泰勒级数展开法证明了被控对象在输入干扰和输出测量噪声均有界的情况下,高阶PID采样迭代学习控制算法的收敛性,并且得出了收敛条件。由于收敛条件中没有积分项,因此更加利于分析计算。与传统的一阶采样迭代学习控制算法相比,高阶采样迭代学习控制算法由于利用了更多先前的控制信息而能使被控对象的实际输出更加接近理想输出。给出了相应的数值仿真,证明了理论分析的有效性。与此同时,结合啤酒生产过程中糖化阶段中酒花添加等实际问题对该算法的应用前景作了一定的分析。
Higher-Order PID Sampled-Data Iterative Learning Control algorithm （HOSDILC） for a class of uncertain nonlinear system with time-delay and disturbances and discusses the advantages and disadvantages of this algorithm is presented. By using Taylr＇s series, it proves the convergence of the HOSDILC. As the mathematical expression for convergence contains no integral term, it is much easier to calculate. Comparing with the traditional one-order sampled-data iterative learning control, HOSDILC has a better control perform- ance as it utilize much more early information. A numerical example is given to prove the efficacy of the theoretical analysis. Mean- while, based on the practical problems such as hop dosing during the beer brewing, the perspective of this algorithm in the practical environment is analyzed.
出处
《控制工程》
CSCD
北大核心
2012年第1期73-76,共4页
Control Engineering of China
基金
国家自然科学基金资助项目(NSFC60974001)
江苏省"六大人才高峰"项目
关键词
高阶PID采样迭代学习控制
时滞非线性系统
泰勒展开
higher-order PID sampled-data iterative learning control
nonlinear system with time-delay
taylor series
