SEMIEMPIRICAL MODELING OF NONLINEAR DYNAMIC-SYSTEMS THROUGH IDENTIFICATION OF OPERATING REGIMES AND LOCAL MODELS

An off-line algorithm for semi-empirical modeling of non-linear dynami c systems is presented. The model representation is based on the inter polation of a number of simple local models, where the validity of eac h local model is restricted to an operating regime, but where the loca l models yield a complete global model when interpolated. The input to the algorithm is a sequence of empirical data and a set of candidate local model structures. The algorithm searches for an optimal decompos ition into operating regimes, and local model structures. The method i s illustrated using simulated and real data. The transparency of the r esulting model and the flexibility with respect to incorporation of pr ior knowledge is discussed.