一种离散系统模型降阶的新方法

A NEW APPROACH FOR MODEL REDUCE ON DISCRET SYSTEM

本文引入了误差多项式的概念,用误差多项式的系数表示原模型与降阶模型之间任意时刻的差异,使原模型与降阶模型两者的Markov参数和H参数部分一致,使得误差多项式的某些系数为零,从而得到降阶模型。本方法简单、灵活、易行,物理概念明确、适合于工程应用。

This paper has introduced a concept of error polynomial. The coefficientsof error polynomial indicate the difference between original model and reduc-tion model at any moment, conform Markov parameter of original model to thatof reduction model, conform theaparttef Hpparameter of original model to thatof reduction model and cause certain coefficients of error polynomial to be zeroso as to obtain the reduction model. This method is sasy, simple and flexible.The physics concept is determinate. It is applicable to engineering application.