Detection and tracking of short duration variations of power system disturbances using modified potential function

Abstract

This study introduces a modified potential function based modeling approach for real-time disturbance tracking. Real-time tracking requires the least complex techniques for processing and classification and still provide accurate results. Time required to detection of the power quality disturbance should be small enough so that the further mitigation action may be taken. The proposed FRTD algorithm is able to detect the disturbances in 3–5 samples. The proposed FRTD algorithm is principally based upon potential function, which has been modified so as to give the one-dimensional position vector corresponding to the voltage sag and swell, with complexity O(1) with no pre-process complexity. The method detects the transient in 3–5 samples at sampling frequency of 12.8kHz. Potential function uses the current nonlinear drift velocity vector along with a prior knowledge of the power system disturbance to compute the unknown parameter in the form of diffusion matrix, Dt. The parameter Dt has been used to detect the events in the signal. Magnitude of Dt play vital role to detect the events. FRTD algorithm has been tested to track and detect the power system disturbances such as voltage sag, swell, and transient. The algorithm has been tested on hardware interfaced with MATLAB. 1000 samples of different types of power system disturbances such as voltage swell, voltage sag, oscillatory-transient and impulsive-transient have been tested. FRTD algorithm is having the efficiency of 99.83%.