A Feasibility Study of the Application of Signal Processing Techniques to Corona Discharge Characterization on HVDC Systems †

Abstract

This paper presents an initial and feasibility study of corona discharge characterization on HVDC systems using an alternative strategy based on digital signal processing techniques. In the corona discharge test, high voltages varying from ±30 to ±80 kV were applied through a DC high-voltage generator. In addition, corona discharges around the conductor were measured by means of a data acquisition prototype system equipped with a metal electrode device for the purpose of corona current measurement. The signals collected were subjected to digital signal processing parameters to extract the most relevant information related to corona discharge occurrence in specific frequency bands. The results reveal that the proposed method was able to detect the corona discharge characteristics following the changes in the signal content, especially, for narrow frequency ranges. The results indicate the feasibility of the proposed method to detect and characterize the corona discharge in a simple way, which expands the research field in corona discharge characterization in HVDC systems by means of digital signal processing and feature extraction.