parametrization for a state space model

3 ビュー (過去 30 日間)
karthik pakala
karthik pakala 2017 年 5 月 15 日
編集済み: karthik pakala 2017 年 5 月 15 日
I have been using a software SIDPAC for aircraft system identification. In the estimation process I need to apply some constraints on the parameters that are being estimated. The state space model that is being used for the output error modelling is as follows,
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%code start
function [Y,X,A,B,C,D] = YawHeavDyn(p,U,w,x0,c)
n=c(1);
A=[p(1),p(2),n;...
p(3),p(4),p(5);...
n,p(6),p(7)];
ns=size(A,1);
B=[p(8),0;...
p(9),p(10);...
0,0];
C=[1,0,0;...
0,1,0;...
0,0,1];
D=[0,0;...
0,0;...
0,0];
[no,ni]=size(D);
nw=length(w);
jay=sqrt(-1);
jw=jay*w;
Y=zeros(nw,no);
X=zeros(nw,ns);
for i=1:nw,
X(i,:)=((jw(i)*eye(ns,ns)-A)\(B*U(i,:).')).';
Y(i,:)=(C*X(i,:).' + D*U(i,:).').';
end
return
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%code end
The parametric constraints that i have to apply are as follows,
p(7)=2*p(1)
p(10)=-p(5)
p(4)=p(3)
p(6)==[-inf 0]>>p(6) is less than zero.
[y7,p11,crb,rr]=fdoe('YawHeavDyn',p10,U,t,w,c,z);%%fdoe method used to estimate the output error model
p10-parameter vector
u-input vector
t-time vector
w-frequency vector
c-constant vector
z-output vector
Thanks in advance

回答 (0 件)

カテゴリ

Help Center および File ExchangeLinear Model Identification についてさらに検索

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by