Short duration power quality disturbances detection and mitigation using custom power devices

Abstract

The widespread usage of power electronic devices has led to a complete change in electrical load nature and these variations in load causes disturbances in voltage waveforms Power quality PQ has gained much of its importance Due to usage of sensitive power electronic instruments some of the disturbances arises leading to reduce the efficiency The major consequences of voltage disturbances are loss of efficiency malfunction and damage of equipment Hence compensation of these disturbances using custom power devices are required for power quality improvement The short duration voltage disturbances such as voltage sag swell and interruption etc draws extensive attention in this thesis The research work starts with the detection and classification of various short duration power quality problems such as sag swell and interruption The disturbances are first classified using Feed Forward Back Propagation Neural Network FFBPNN algorithm The synthetic training samples are taken from the horizontal and vertical histogram using simulation These samples are used to train the FFBPNN algorithm for the classification of power quality problems The second approach focuses on the detection and classification of the voltage disturbances using Discrete Wavelet Transform DWT signal processing technique The combination of these features are computed using Multi Resolution Analysis MRA and Daubechies Db4 DWT Next work is the compensation of the short duration voltage disturbances The use of custom power devices such as Dynamic Voltage Restorer DVR is the most efficient method for restoration of system voltage newline