Improving frequency and ROCOF accuracy during faults, for P class Phasor Measurement Units

Andrew Roscoe, Graeme Burt, Gert Rietveld

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

7 Citations (Scopus)
60 Downloads (Pure)


Many aspects of Phasor Measurement Unit (PMU) performance are tested using the existing (and evolving) IEEE C37.118 standard. However, at present the reaction of PMUs to power network faults is not assessed under C37.118. Nevertheless, the behaviour of PMUs under such conditions may be important when the entire closed loop of power system measurement, control and response is considered. This paper presents ways in which P class PMU algorithms may be augmented with software which reduces peak frequency excursions during unbalanced faults by factors of typically between 2.5 and 6 with no additional effect on response time, delay or latency. Peak ROCOF excursions are also reduced. In addition, extra filtering which still allows P class response requirements to be met can further reduce excursions, in particular ROCOF. Further improvement of triggering by using midpoint taps of the P class filter, and adaptive filtering, allows peak excursions to be reduced by total factors of between 8 and 40 (or up to 180 for ROCOF), compared to the C37.118 reference device. Steady-state frequency and ROCOF errors during sustained faults or unbalanced operation, particularly under unbalanced conditions, can be reduced by factors of hundreds or thousands compared to the C37.118 reference device.
Original languageEnglish
Title of host publicationProceedings of the 2013 IEEE International Workshop on Applied Measurements for Power Systems (AMPS)
ISBN (Print)978-1-4673-5571-1
Publication statusPublished - Sep 2013
Event2013 IEEE International Workshop on Applied Measurements for Power Systems (AMPS) - Aachen, Germany
Duration: 25 Sep 201326 Sep 2013


Conference2013 IEEE International Workshop on Applied Measurements for Power Systems (AMPS)


  • fourier transforms
  • frequency measurement
  • filtering algorithms

Cite this