Optimal parameter estimation of synchronous machines from the digitized sudden short-circuit armature current

This paper presents a novel approach for the optimal estimation of the nonsaturated parameters of synchronous machines from the digitized samples of the sudden short-circuit armature current. The proposed technique uses the least error squares (LES) parameter estimation algorithm to identify these parameters. The proposed technique splits the armature current into three currents, namely; the steady state armature current, the transient current, and the subtransient current. The parameters of each current, namely, x/sub d/,x/sub d/'T/sub d/' and x/sub d/" and T/sub d/" are determined. After identifying these parameters, the algorithm uses them to determine the digitized samples of the DC armature transient current, simply by subtracting the armature current from the constructing armature current using these parameters. These samples are then used to identify the armature time constant T/sub a/ as well as the subtransient reactance X/sub q//sup N/. The proposed algorithm is tested using simulated data. Effects of the critical parameters on the algorithm are also studied in this paper.