Matrix Pencil method as a signal processing technique performance and application on power systems signals

Abstract

The performance of Matrix Pencil Method (MPM) has been analyzed and a new application in sub-cycle fault signal analysis using MPM has been proposed and tested. The performance of MPM in relation to sampling window width, sampling frequency and damping factor has been statistically analyzed in the first part of the thesis. In the second part of the thesis, MPM has been applied to realistic fault signals simulated in the IEEE 34-bus test system to classify the fault type based on feature extraction of space vectors and zero-sequence signals. It was found that while using MPM alone was able to provide a correct fault classication using 15 ms of post-fault data, augmenting an ellipse tting algorithm to MPM could improve the performance the fault classication to using 5 ms of post-fault data.