A discriminative approach to harmonic emission assessment

PhD (Electrical Engineering), North-West University, Potchefstroom CampusHarmonic emission assessment of a Renewable Power Plant (RPP) for grid code compliance, is a complex task that requires a fair and pragmatic approach. RPPs should only account for the harmonics they emit, i.e. those which negatively affect the harmonic voltage at the Point of Connection (PoC). Most of the harmonic emission assessment methods require the use of measured harmonic phasors. However, harmonic phasor measurements taken for more than 24 hours, result in an enormous volume of data that is cumbersome to analyse. Power Quality (PQ) measurements is typically done over days (in some cases over 7 days [1]), which is why power quality data is aggregated to reduce the volume of data. Aggregation of rms data is a known concept that is detailed in the IEC 61000-4-30 power quality measurement standard. Depending on the aggregation time interval selected, datasets of 3-sec, 10-mins or 120-min time intervals can result. The challenge is that only rms data are retained in this standard, with phase angle information being discarded. With the constraint of using a 10-min rms value, the CIGRE/CIRED joint working group C4.109 developed a statistical approach that makes use of harmonic voltage and harmonic current scatterplots with reference impedances of the upstream network and the downstream nonlinear load to qualitatively evaluate which party is the major contributor towards the harmonic distortion measured at the Point of Connection1 (PoC). The CIGRE/CIRED C4.109 statistical method (in this thesis) was found to be inadequate in conclusively indicating which harmonics are being emitted or being absorbed by a PV plant when only one measurement point is considered at the PoC. By application of synchronised and coherent measurements at multiple points, the method was improved to discriminate harmonic emission within the 10-minute rms constraint. RPPs or nonlinear loads may not be able to conduct a multiple measurement point campaign, as they only have access to their PoC. Therefore, an alternative approach is required to evaluate their harmonic emissions. Aggregation of harmonic phase angles, in line with IEC61000-4-30 principles, is a new concept that results in a reduced dataset consisting of aggregated harmonic phasors, that can be used to continuously assess harmonic emission at a single measurement point. Application of aggregated harmonic phasors to a PV plant undergoing grid code compliance, found that the discrimination of harmonic emission at a single measurement point can be done through the Range of Secondary Cancelation (RSC) technique, which has thus far only been discussed in theory. Confidence in the RSC technique and aggregated harmonic phasors were further confirmed with field measurements (over a 60-day period) at an industrial site with the aim to monitor harmonic filter performance. It was found by application of the RSC technique, that one of the filter banks did not perform as designed, while the other filter banks effectively reduced the harmonic voltage at the PoC. The RSC technique achieved discrimination of harmonic emission values, however the actual harmonic emission value used for determining compliance to emission limits is to be evaluated. Application of the IEC 61000-3-6 general summation law was found to be too conservative due to the use of summation factors in the summation of harmonic distortion. In a real network harmonic distortion summation is dynamic especially when other non-linear loads are connected to the Point of Common Coupling (PCC). The “prevailing phase angle” value was used as a representative value of the harmonic dataset to result in a more accurate harmonic emission assessment in this integrated network.Doctora