Identification of a Nonlinear Errors-in-Variables Model

Abstract In this paper a nonlinear MISO model is built up as a combination of a pre-specified dynamical term and an unknown static nonlinear component, where the nonlinear subsystem is realized by a region-wise base point driven interpolation. A model using noisy observations (errors-in-variables) is considered. The paper presents the identification algorithm of this model and shows the application of the elaborated method for load forecasting.