The impact of network changes on power quality and compensation device perfomance.

Abstract

This dissertation describes the impact of changing network configuration on power quality and performance of existing compensation devices in the transmission network. The underlying theory was assessed and thereafter the Everest substation network case scenario was selected to study the above due to; the number of reconfigurations it has experienced in the past, increased capacitor bank failures and also due to the harmonics problems experienced. The study involved the installation of harmonic current measuring instruments at Everest 132kV feeders to identify the potential sources of harmonics and to determine the dominant harmonics. A dig Silent Power Factory model was then constructed to perform various simulations in order to determine the impact of the changes done on the Everest network as well as the impact of capacitor switching on the harmonics amplification at Everest 132kV Bus-bars. The study also focussed on analysing the performance history of the capacitor banks at Everest and to determine if high harmonic amplification had an impact on capacitor bank performance. The simulation results revealed that network reconfigurations have negatively impacted power quality at Everest. The results showed that there is also a correlation between the switching of the two 72MVAR capacitor banks and the amplification of the harmonics at Everest. The highest amplification occurred when both capacitor banks were switched in and the resonance point occurred around the 5th harmonic which coincided with the data from field measurements. There was a 61% increase in 5th harmonic impedance amplitude after the Everest network was reconfigured, for the condition when both capacitors are switched in. The lowest amplification occurred when none of the capacitor banks were switched in. Three options were assessed to eliminate the problem of harmonics at Everest, the first one was to prohibit the switching in of both capacitors at Everest and utilise other available means around the Everest network for voltage support. The second option was to change the capacitor size, thus moving the resonance point away from the 5th harmonic. The last option was to install a harmonic filter at Everest to filter out the problematic harmonics. The first option is recommended as it has been successfully tested, can be readily implemented and is much more cost effective compared to the others.