Improvement of cable defect identification for power distribution networks

Abstract

Power distribution networks are critical infrastructures that secure delivery of electricity in a power system. Since electricity is mainly distributed through underground 11 kV and 22 kV power cables on Hong Kong Island, the maintenance of underground power cables is one of the most challenging jobs that The Hongkong Electric Co., Ltd. (HK Electric) has to face. This is because excavation for repair and replacement is difficult and time-consuming due to busy traffic and congested underground conditions. In order to improve maintenance efficacy, this study puts forward a new method to screen and locate potential defective component(s) of underground power cables by collectively using online and offline partial discharge (PD) diagnosing methods. The study also introduces new evaluation metrics (standard deviation of mean tan delta, and delta of mean tan delta) to identify cables at risk of developing water trees inside the cross-linked polyethylene (XLPE) insulation. The effectiveness of these two proposed methods has been validated on its distribution network. These improved identification methods for detecting cable defects could further enhance the reliability and security of HK Electric’s power distribution system.