Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
Drexel University or The Math Forum.
|
|
|
|
StateSpace in simulink
Posted:
Sep 20, 2007 1:25 PM
|
|
Hi folks,
I have a simple - maybe stupid - question !?
I have used the system identification toolbox to identify a statespace-model with multiple inputs and a single output (MISO). I now want to use this ss-model in simulink ....
The model is a continuous model with sampletime = 1, I use the workspace model's parameters (model.A, model.B etc.) in my simulink statespace-block, set simulation configurations to fixedstepsize=1 (ODE5) and reconstruct my input vector according to the one I used for identification with a mux block ..... but simulation doesn't work, the output diverges ...
I'm not that much of a simulink user, but the model itself is not unstable when I use it with matlab (sim, step, compare etc.) ..... What am I doing wrong, do I have to regard something about the configuration parameters, stepsizes, do I have to choose another block (discrete statespace ??) ... ?? I'm a little confused now, should be possible to use the model in simulink, so it must be a simple mistake ..... but I don't get it ... Glad if someone could help ...
Andreas
|
|
|
|