Validation and improvement of Models in the Frequency Domain

Classical validation methods "accept" or "reject" a model as a valid representation of a plant. However this binary result has several problems: i) Models are neither good nor bad but have a certain valid frequency range, ii) No indication for reasons of failure or how to improve the model, iii) In iterative identification and control approaches undermodeling is usually present. This fact makes it difficult to apply traditional model validation schemes. We present a new validation procedure that overtakes these problems by performing the validation in the frequency domain. Hence the validation/invalidation process is frequency dependent permitting to ascertain for which frequencies the model is not validated. Moreover this information is used as a guideline to find a more appropriate model structure