Simulation with non linear Hammerstein-Wiener models
3 views (last 30 days)
Show older comments
I trained a non linear Hammerstein-Wiener model.
To test it i had Data = iddata(OutputData,InputData,1) and i used the command sim(model,InputData) but the results are very different from the output, but if i do compare(Data,model) the fit is very good and the result is close to the output.
Then i found that i was missing the states so by doing sim(model,InputData,findstates(model,Data)) i obtain the same result as compare.
Now my problem is that i want to use this model in "real-time" so giving to the model the first value of InputData calculate the output with sim then give the second value and so on, but i don't know how to do it since i don't have OutputData at the start to calculate the states. (for example i tried with a nlarx model but in this case is simpler because i need only to pass the regressors)
I also found this example How the Software Computes Hammerstein-Wiener Model Output to avoid the using sim but i have the sme problem, i guess that in this example ysim is correct because initial conditions is zero so it is not specified and don't cause problem.
Thank you for the attention, hope the problem is clear.
0 Comments
Answers (0)
See Also
Categories
Find more on Transfer Function 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!