Applying extended Kalman filter with moving horizon to fault estimation in power system

Assem Thabet (1) , Mohamed Boutayeb (2) , Gaëtan Didier (3) , Saïd Chniba (1) , Naceur Abdelkrim (1)

1 ENIG - Ecole Nationale d'Ingénieurs de Gabès
2 CRAN - Centre de Recherche en Automatique de Nancy
3 GREEN - Groupe de Recherche en Electrotechnique et Electronique de Nancy

In this paper, a useful observer-based fault estimation (FE) method is presented for power system. The nonlinear dynamic model based on a differential algebraic equation (DAE) is transformed to a simple ordinary differential equation (ODE). Two nonlinear observers are used for FE for comparison purposes, one of them being an extended Kalman estimator and the other is a state observer with moving horizon with a study of convergence based on the choice of the matrix of covariance of the noises of system and measurements. Some simulations studies are carried out on the IEEE 3 buses test system.

Domaines: Automatique / Robotique

Assem Thabet, Mohamed Boutayeb, Gaëtan Didier, Saïd Chniba, Naceur Abdelkrim. Applying extended Kalman filter with moving horizon to fault estimation in power system. 12th international conference on Sciences and Techniques of Automatic Control and Computer Engineering, STA'2011, Dec 2011, Sousse, Tunisia. pp.CD-ROM. ⟨hal-00651144⟩