Tools for Semi-physical Modeling

By semi-physical modeling we mean such an application of system identiication, where physical insight into the application is used to come up with suitable nonlinear transformations of the raw measurements, so as to allow for a good model structure. Semi-physical modeling is less "ambitious" than physical modeling, in that no complete physical structure is sought, just suitable inputs and outputs that can be subjected to more or less standard model structures, such as linear regressions. In this contribution we discuss various tools that can support the process of semi-physical modeling. We will deal both with analytical tools such as diierential algebra and more informal ones such as the programming environment. 1. INTRODUCTION System identiication is applied to a wide variety of processes. Loosely speaking the idea is to t a model structure to observed data, so as to come up with a model which as well as possible reproduces the past behavior of the system. The crux in this process is really to nd a suitable model structure for the process and for the intended application of the model. Several approaches to the design of such a model structure have been described and discussed earlier in the literature: