摘要
本文针对大系统的频域模型简化提出了两种方法:最小二乘近似法和有理分式近似法.前者利用了原系统传递函数G(s)展开式幂级数中远远大于2m(m为简化模型的阶次)项的信息来确定简化模型的参数;后者直接用原系统传递函数的参数求得简化模型的参数.与帕德近似法不同的是:它们充分利用了原模型的信息,提高了模型简化的精度.
The two methods of model simplification are presented for large scale systems, they are the least square approximation and the rational fraction approximation. The former determines the parameters of the simplified model by using the many items of the series of the original transfer function G(s) at s = 0 ,and the latter does it directly by using the parameters of GO) . They have higher precision in comparison with previous methods,for example Pade method.
出处
《系统工程》
CSCD
1998年第5期60-66,共7页
Systems Engineering
关键词
大系统
频域
模型简化
模型降价
Large scale system, model simplification, model reduced-order
