Power Quality Disturbances and Protective Relays

Abstract: This thesis combines knowledge from two areas: power quality and power system protection. Power quality concerns the effect of voltage disturbances on end-user equipment and of current disturbances on network components. Power system protection concerns the detection of faults in the system from measured voltages and currents. However, non-fault voltage and current disturbances may lead to the inadvertent detection of a fault in the same way as voltage disturbances may lead to the tripping of end-user equipment.The aim of the research presented in this thesis is to quantify non-fault voltage and current disturbances at the terminals of protective relays; in other words: to define power quality from the viewpoint of a protective relay.Starting from the principles (algorithms) of the relays, a quantification approach is developed to assess the disturbance impact on the relays. The quantification is based on the relay response after processing the disturbance signals by relay filters. This makes it possible to quantify the disturbance impact by the consequences caused by the disturbances, other than by the signal waveforms of the disturbances.For component switching disturbances, the so-called "impact severity curve" is developed to quantify the disturbance impact on a relay. It is a measure of the potential risks that a relay faces in case of a particular power system disturbance. It also facilitates the relative comparison among different disturbance cases. For frequency deviation disturbances, the resulting errors in relay outputs are estimated, which is helpful in determining whether an accurate frequency measurement is needed.The proposed approach in the project can be applied in testing the disturbance immunity of any future fast relay algorithm. The test results, which are in the form of impact severity curves, can be adopted as the basis of disturbance classification in the viewpoint of protective relays and further lead to the setup of immunity test database.