Advanced techniques for on-line PD measurements in high voltage systems

F. Álvarez,J. Ortego,F. Garnacho,M.A. Sánchez-Urán,F. Alvarez,M.A. Sanchez-Uran

DOI: https://doi.org/10.1109/ichve.2014.7035404

2014-09-01

Abstract:Partial discharges (PD) measurements provide valuable information for the condition assessment of the high voltage insulation systems. PD tests have become a fundamental tool to determine the status of high voltage elements, contributing to the quality assurance in high voltage equipment. Some reasons for the increasing usage of DP tests are based in the effectiveness of the new developments presented to achieve more reliable diagnosis. During the last five decades different approaches have been developed and applied to PD measurement and processing techniques. Significant efforts have been made to deal with the most common problems arising in on-site and on-line measurements: high levels of noise and interferences, the emergence of different PD sources simultaneously and the difficulty of locating the site of the defects and to identify the type of defects. In order to overcome all these drawbacks, this paper proposes an efficient method based in advanced measuring and diagnosis techniques, with the aim to achieve a proper sensitivity in PD measurements and an accurate assessment of the elements insulation condition. In PD measurements the main factors that condition the signal acquisition, the subsequent processing and the capability to obtain an accurate diagnosis are: selection of a suitable detection technique together with the type of sensors and their bandwidth, implementation of an effective noise filtering tool, development of a defect localization method and implementation of PD classification techniques for pattern recognition and defect identification. All these factors have been considered in this study and a complete measuring method together with powerful diagnostic tools are presented. Measurements in a test object consisting in a one-phase cable system were accomplished with real insulation defects, with the aim to show the effectiveness of the proposed method and the optimized analysis techniques.