Unknown-input observers (UIOs) allow for estimation of the states of an LTI system without knowledge of all inputs . We provide a noveldata-driven UIO based on behavioral system theory and the result known asFundamental Lemma proposed by Willems et al. We give necessary andsufficient conditions on the data collected from the system for the existence of a UIO providing asymptotically converging state estimates, and propose apurely data-driven algorithm for their computation . We apply the proposed method to distributed state estimation inDC microgrids and illustrate its potential for cyber-attack detection .

Author(s) : Mustafa Sahin Turan, Giancarlo Ferrari-Trecate

Links : PDF - Abstract

Code :

Keywords : data - system - state - estimation - driven -

Leave a Reply

Your email address will not be published. Required fields are marked *