用行搜索算法的最小实现方法及CAD

本实现方法是通过马尔柯夫参数构成适当阶数的 Hankel 矩阵,应用行搜索算法搜索其线性独立行和计算其线性相关行的线性组合系数,从而求得系统的最小实现。其优点是不管传递函数阵 G(s)是否已经约简,用本法求得的实现总是最小实现。文中介绍了该法的计算机辅助设计问题,其计算并不复杂,编制程序也较容易。此外,文中还提出了一种新的极点配置算法。应用这种新算法,则用本法所求得的最小实现,不必再进行任何其他变换,即可直接用于状态反馈。

Using Row-searching Algorithm to Find Minimnm Realization and Its CAD

A method is proposed of using Markov parameters to form a Hankel matrixof appropriate orders and of applying the row-searching algorithm to search itslinearly independent rows and to calculate the linear combination coefficients ofinearly dependent rows.Thus,a minimum realization of the system can be obtained.The advantage of this method lies in the fact that the realization is always minimumno matter whether the transfer function matrix G(s)is irreducible or not.Thecomputer-aided design of this method is also introduced.The calculation involvedis not very complicated and programming is easy.In addition,this paper presentsa new pole-assignment algorithm by which minimum realization can be applieddirectly to state feedback and there is no need for any other transformation.