Diagnosis of EHV and HV Transformers Through an Innovative Partial-Discharge-Based Technique

Online partial discharge (PD) analysis of power transformers is sometimes complex due to the presence of large disturbance and noise as well as to uncertainty in PD source identification. In this paper, a novel approach to PD detection and analysis in online power transformers is presented. The proposed approach provides hardware and software tools to remove disturbance, noise, separate contributions from different sources and, eventually, identify PD sources. Besides, a description of the hardware tools used to assemble the diagnostic system, the tests (in the lab and on real transformers) carried out to build the reference database needed for identification system training will be discussed, providing rules for PD source identification. A number of working cases will be presented to show the effectiveness of the proposed method.