DIFFER Seminar: Improving Baseline Models through Model Augmentation: Model Structures and Learning Approaches

Abstract: In most engineering applications, strong prior knowledge is present in the form of pre-existing models, provided by system designers and engineers, even though they do not capture all the nonlinear dynamics of the real-life system. These models are currently not accounted for during black-box system identification / data-driven modelling tasks. We aim to develop a comprehensive data-driven modelling framework to obtain accurate and interpretable models of measured complex system dynamics by augmenting an approximate pre-existing model through black-box nonlinear system identification. During this talk we will explore new theory and algorithms to provide model structures, algorithms and theory that flexibly interconnect the pre-existing model and the black-box completion 2