I have constructed a sate space model of a non linear plant and would like to evaluate the 'correctness' of said model compared to the real plant. As for now I have logged data for all states from the plant and use that as an input to my model and integrate the output ($\int\dot{x}dt$) to get get calculated state values. I then compare the two data series in a plot. Though this gives a fair feeling for how 'well' the model is compared to the real plant, I would like to be able to have a more quantifiable answer of how 'good' the model is.

My question is what methods are used to evaluate the performance of a model compared to the real plant, and what litterature exists on the subject.



You must log in to answer this question.

Browse other questions tagged .