Hi i have estimated an armax model using python sippy library. The estimation gives me two transfer functions H and G. How can I combine them into a single one to predict model output for new input u(t)or to compute unit step response? I thought to somehow derive state space representation maybe...
André Kummerow In your case, you've estimated two transfer functions, H and G, using the Python SIPPY library. These transfer functions represent different aspects of the ARMAX model:
Now, to predict the model output for new inputs or to compute the unit step response, you need to combine these transfer functions. Here's a simple way to understand the combination process:
Thank you very much for your well-founded answer. Sounds reasonable. Then I just have to figure out, how SIPPY approximates the noise signal, since this is not defined as separate input signal, when estimating H and G. It seems to be somehow calculated internally. But the noise is not related to u(t) and not y(t) or the prediction error of y(t)?
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running