How to realize a nlarx system in Simulink without using the Toolbox block
1 view (last 30 days)
Show older comments
Hey all,
I am using System Identification Toolbox for constructing non linear ARX models. I already got nice nonlinear ARX systems which I named "motor_model_left" and "motor_model_right".
They have 6 nonlinear regressors: y1(t-1), y1(t-2), y1(t-3), u1(t), u1(t-1), u1(t-2).
I sent the systems to the workspace.
I also know, that nonlinear systems can be understood as an extension of a linear model. A linear SISO ARX model has this structure:
y ( t ) + a_1 y ( t - 1 ) + a_2 y ( t - 2 ) + ... + a_na y ( t - na )
= b_1 u ( t ) + b_2 u ( t - 1 ) + ... + b_nb u ( t - nb + 1 ) + e ( t )
(e(t) can be excluded)
My problem now is, that I don't know how to get the a and b coeffitients.
I need them because I have to create a Simulink model to simulate the motors without using the System Identification Toolbox blocks (in this case the Nonlinear ARX Model).
I already tried "motor_model_left.Nonlinearity.Parameters.LinearCoef"
and got: 0.3714, 0.0192, -0.0039, -0.0816, -0.0066, 0.0068 .
So I built a Simulink model seen in the png.
But that thing doesn't work. Either the coeffitients are wrong or my model.
Can somebody help me?
Thanks a lot!
0 Comments
Answers (0)
See Also
Categories
Find more on Nonlinear ARX Models in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!