📄 stdc.m
字号:
function [K,Kf,P,ro,Tx]=stdc(G,plt)
% [K,Kf,P,ro,Tx]=stdc(G,plt) inversion based static decoupling control.
% Given a plant G (mutools format), this function provides a control
% law which moves a pole on every zero, and moves in zero the
% other poles of G.
% K and Kf are the feedback and feedforward matrices, such that
% the closed loop system is P(s)=(C-D*K)*(s*I-A+B*K)*B*Kf+D*Kf
% Every output y(i) of P(s) is the ro(i) derivative of ist input v(i),
% where ro(i) is the relative degree of y(i).
% Tx is the transformation matrix which brings the system in normal
% form : An=Tx*A*inv(Tx), Cn=C*inv(Tx), Bn=Tx*B; Dn=D; see statecc.
%
% If plt=1 (default) the function produces plots and displays data.
% Default w is a 100 log spaced points vector from 1e-6 to 1e6 rad/sec.
%
% Example :
% A=rand(3);B=rand(3,2);C=rand(2,3);D=rand(2,2);
% G=[[A B;C D] [3;zeros(4,1)] ; [zeros(1,5) -inf]];
% [K,Kf,P,ro,Tx]=stdc(G);
%
% G.Campa 22/1/98
if nargin<2,plt=1;end
% general info
[ty,no,ni,n]=minfo(G);
if ty=='cons', G=[G,zeros(size(G,1),1);zeros(1,size(G,2)),-inf]; end
[A,B,C,D]=unpck(G);
Tx=[];
ro=zeros(no,1);
Ex=zeros(no,n);
Dx=zeros(no,ni);
% relative degree
for i=1:no,
for k=0:n-1,
if k==0,Dx(i,:)=D(i,:); else Dx(i,:)=C(i,:)*A^(k-1)*B;end
if any(Dx(i,:)~=zeros(1,ni)),
ro(i)=k;
Ex(i,:)=C(i,:)*A^k;
break,
else
Tx=[Tx;C(i,:)*A^k];
end
end
end
Tx=[Tx;null(Tx)'];
Kf=pinv(Dx);
K=Kf*Ex;
P=pck(A-B*K,B*Kf,C-D*K,D*Kf);
%------------------------------------------------------------------------%
% poles drifting:
if plt,
figure;
pzmap(A-B*K*0,B*Kf,C-D*K*0,D*Kf);
hold on
title('closed loop poles drifting')
avm=[];
for q=0:.02:1,avm=[avm eig(A-B*K*q)];end
l=size(avm,2);
for j=1:size(avm,1),
plot(real(avm(j,1:l-1)),imag(avm(j,1:l-1)),'r.')
plot(real(avm(j,l)),imag(avm(j,l)),'bx')
end
end
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -