Covariance Based Estimation For Reduced Order Models Of Microgrid Power Flow Dynamics

Identification using system data becomes a great tool for modeling a system when model derivation using physical laws is complex or impossible. Covariance Based Realization Algorithm (CoBRA), one kind of subspace methods for system identification, allows the use of finite size covariance data matrices to reduce storage requirement. In addition, the covariance pre-processing reduces noise effect and allows CoBRA to concentrate on the identification of possibly low order deterministic dynamics. This paper aims at providing practical analysis of CoBRA in the area of power systems. This includes persistent excitation, identification from arbitrary data segments, the effect of sample approximation and imperfect data. Finally, an application of CoBRA to the identification of microgrid power flow dynamics is demonstrated using available synchrophasor measurements. (C) 2018, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.