Abstract
The volume of data produced by existing partial discharge monitoring systems is often too large for engineers to examine in detail, leading to data being ignored and useful indicators of health being missed. The case study reported in this paper recorded 21\thinspace839 events around an HVDC reactor over a six-day period. We estimate that it takes 1 min to check whether an event requires detailed study, leading to over two man-months of effort to locate important events in a dataset of this size. Additionally, online monitoring data are stored onsite, and may require an engineer's visit for collection. This paper presents an approach to remote partial discharge monitoring, supported by automated data interpretation and prioritization, which enables engineers to remotely find and download important data. Results from the case study are used to illustrate these concepts.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
