系统辨识中最优采样间隔的确定  

Determination of optimal interval in system identification

作　　者：李立刚[1,2] 张召昕[2] 戴永寿[2] 张亚南[2] 昂扬[2] 

机构地区：[1]北京科技大学自动化学院,北京100083 [2]中国石油大学(华东)信息与控制工程学院,山东东营257061

出　　处：《电子设计工程》2013年第11期24-26,30,共4页Electronic Design Engineering

摘　　要：针对如何确定连续系统辨识中最优采样间隔的问题,根据复平面中辨识模型的特征值到原点的距离判断采样间隔是否合理,利用折半查找算法,设计了一种在给定的范围内寻找系统辨识中最优采样间隔的方法。该方法可在辨识对象的特征值未知的条件下得到采样间隔的最优值,使辨识所得模型的输出误差最小。最后,文中通过仿真实验验证了该方法的有效性。Input and output signals should be sampled before the identification of a continuous system. To solve the problem of how to determine the sample interval, this paper proposed a method for searching the optimal sample interval based on the existing method which uses the distance between the original and the characteristic value of the identified model, combined with the binary search. The optimal sample interval which minimizes the output error of identified model can be obtained by using this method without knowing the characteristic value of the original system, and the simulation result verified the efficiency of the novel method.

关 键 词：系统辨识 采样间隔 折半查找 辨识误差 

分 类 号：N945.14[自然科学总论—系统科学]