航空学报 > 1980, Vol. 1 Issue (1): 82-91

利用“观测器”设计数字滤波网络

张伏中   

  1. 北京机电工程设计部
  • 收稿日期:1979-01-01 修回日期:1900-01-01 出版日期:1980-06-25 发布日期:1980-06-25

DIGITAL FILTER NETWORK DESIGN BY USING OBSERVERS

Zhang Fuzhong   

  1. Beijing Electrical and Mechanical Engineering Institute
  • Received:1979-01-01 Revised:1900-01-01 Online:1980-06-25 Published:1980-06-25

摘要: 本文介绍利用“观测器”设计数字滤波网络的方法。本方法是先给出系统要求的闭环对象极点位置,然后求得满足系统特性的数字滤波网络;而惯用的对数频率法则是基于系统的开环对数频率特性,凭借设计者的经验选取某个网络,用工程作图法使系统特性去逼近希望的理想特性。显然,前者依赖于数字模拟计算,着重考虑网络的数字计算机实现;而后者则取决于设计者的经验,着重考虑网络的物理实现。 另外,本方法对问题的描述采用了状态表示法和矩阵矢量形式,便于数字网络设计的数值计算程序化,因而也就便于修改设计或跨型号的重新设计。

Abstract: A method of digital filter network design using observers is presented. This method consists in providing the required close loop system pole positions, and obtaining a digital filter network satisfying the characteristics of the system; whereas the conventional logarithmic frequency method is based on the open loop logarithmic frequency properties, in which the choice of the network depends on the experience of the designer in making system's properties approach the desired ideal properties by way of engineering graphical curves. The former method depends on digital simulation computation and lays stress on realization of the network design on a digital computer, but the latter method relies on designer's experience and lays stress on physical realization of the network design.State vector and matrix vector representations are used in the numerical program calculation for digital network design through the computer. It makes modification and re-design work easier.