Fault Type Estimation in Power Systems

This paper presents a novel approach for fault type estimation in power systems. The Fault type estimation is the first step to estimate instantaneous voltage, voltage sag magnitude and duration in a three-phase system at fault duration. The approach is based on time-domain state estimation where redundant measurements are available. The current based model allows a linear mapping between the measured variable and the states to be estimated. This paper shows a possible for fault instance detection, fault location identification and fault type estimation utilizing residual analysis and topology error processing. The idea is that the fault status does not change measurement matrix dimensions but changes some elements of the measurement matrix. The paper addresses how to rebuilt measurement matrix for each type of faults. The proposed algorithm is shown that the method has high effectiveness and high performance for forecasting fault type and for estimating instantaneous bus voltage. The performance of the novel approach is tested on IEEE 14-bus test system and the results are shown.