Finite–time identification of the Thévenin equivalent parameters in power grids

Arnaldo Arancibia (1) , Carlos Soriano–rangel (2) , Fernando Mancilla–david (2) , Romeo Ortega (3) , Kai Strunz (1)

1 TU - Technical University of Berlin / Technische Universität Berlin
2 University of Colorado [Denver]
3 L2S - Laboratoire des signaux et systèmes

This paper presents a novel algorithm for online identification of the Thévenin equivalent parameters in ac three–phase power grids of inductive nature. One unique property of the algorithm is to ensure convergence of the estimated parameters within finite time, provided that a weak excitation condition—necessary for identifying offline parameters—is verified. After a finite time, the algorithm reduces to a standard recursive least–squares algorithm to track possible changes in the parameters. Moreover, the algorithm requires neither circuit excitation nor active manipulation of the circuit. The proposed algorithm is evaluated via computer simulation and experimental prototyping for a case study consisting of a three–phase inverter connected to a mains power source via an inductive circuit. Results show that the proposed method is able to successfully identify the Thévenin equivalent parameters.

Domaines: Sciences de l'ingénieur [physics]

Arnaldo Arancibia, Carlos Soriano–rangel, Fernando Mancilla–david, Romeo Ortega, Kai Strunz. Finite–time identification of the Thévenin equivalent parameters in power grids. International Journal of Electrical Power & Energy Systems, 2020, 116, pp.105534. ⟨10.1016/j.ijepes.2019.105534⟩. ⟨hal-03387981⟩