I got the model which is 1/s^2+s+1 I used c2d to make it as discrete domain
y(k)=a1y(k-1)+a2y(k-2)+a3u(k)+a4u(k-1)+a5u(k-2)
%y(k)—the output of the system of k time
%u(k)—the control input of K time
%ai —known constants of System model
X = [x1,x2,x3,x4,x5]'
I try to compute the out put of hidden layer of RBF NN which is
hj = exp(-(( norm(X-cj))^2./bj^(2)));
Please kindly let me know how solve this
Thank you

2 commentaires

Azzi Abdelmalek
Azzi Abdelmalek le 17 Avr 2014
What is X?
Seyed Mohammad
Seyed Mohammad le 17 Avr 2014
Modifié(e) : Image Analyst le 17 Avr 2014
Where:
xi=y(k-i) i=1,2,3
xj=u(k- j +3) j=4,5,6
then
X=[x1,x2,x3,x4,x5]

Connectez-vous pour commenter.

 Réponse acceptée

Greg Heath
Greg Heath le 19 Avr 2014
I do not understand what you are doing.
The obvious model to use is a narxnet with no hidden layers:
net = narxnet(0:2,1:2,[]);
help narxnet
doc narxnet
Search on
greg narxnet
Hope this helps.
Thank you for formally accepying my answer
Greg

Plus de réponses (0)

Community Treasure Hunt

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

Start Hunting!

Translated by