Towards the Character and Challenges of Partial Discharge Pattern Data Measured on Medium Voltage Overhead Lines

Abstract

To detect a fault behavior on medium voltage overhead lines is the key motivation of our research that produced several different metering devices, various long-time real deployments and many academic contributions and experiments. This paper aims to introduce the largest publicly shared dataset having more than 25000 different labeled signals. In the previous phase of our research, we were able to host an international data science based competition on a Kaggle platform while this data sharing tents to invite even more researchers and scientists. We expect that such a data share will open more possibilities into a deeper analysis and wider range of applications.