Feature Extraction for Machine Learning Algorithms to Classify Partial Discharge under DC Voltage

Research output: Contribution to conferencePaper

Abstract

For a safe supply of electrical energy, it is essential to monitor the high-voltage equipment of the power grid and to determine its condition. One of the most important diagnosis methods are partial discharge (PD) measurements. This method and the interpretation of the results are internationally established in the field of AC voltage. Due to current changes in the field of energy supply, a trend away from the classic transmission with AC voltage towards a transmission with DC voltage can be observed. These changes make it necessary to establish PD measuring methods also for DC voltage. However, many of the interpretation methods cannot be adopted, which is why new methods have to be developed. The interpretation of the results can be automated with the help of machine learning, which has many advantages, especially for a continuous PD monitoring. Such systems are already available for AC voltage but not yet for DC voltage. The most important steps of machine learning are feature extraction and analysis. Features are extracted from the raw data and are used to classify and interpret the measurement data. These features must be significant and the different PD defects must be distinguishable from each other by means of these features.
Original languageEnglish
Publication statusPublished - 27 Oct 2020
Event8th International Conference on Condition Monitoring and Diagnosis - Virtuell, Thailand
Duration: 25 Oct 202028 Oct 2020

Conference

Conference8th International Conference on Condition Monitoring and Diagnosis
Abbreviated titleCMD2020
CountryThailand
CityVirtuell
Period25/10/2028/10/20

Fingerprint Dive into the research topics of 'Feature Extraction for Machine Learning Algorithms to Classify Partial Discharge under DC Voltage'. Together they form a unique fingerprint.

Cite this