A weighted-least-squares parameter estimator for synchronous machines

Some applications of synchronous machine models with a desired structure require parameter values which are frequently not readily and satisfactorily obtainable from design data or field tests. An iterative algorithm based on weighted-least-squares estimation theory is proposed to determine directly all standard parameters for a hypothesized model structure by repeatedly processing the armature and field current data of a sudden short-circuit test. Results for a simulated salient-pole synchronous machine with damper winding are included, where the parameters X<inf>d</inf>, X'<inf>d</inf>,X''<inf>d</inf>,X<inf>l</inf>, X<inf>q</inf>, X''<inf>q</inf>, R<inf>a</inf>, T'<inf>do</inf>,and T''<inf>qo</inf>are readily identified in a few iterations with high accuracy.