Electrical Tree Initiation Prediction in Thermally Aged DC XLPE for Cable Insulation
Yi Zhang,Zaijun Wu,Dongdong Zhang,Jiahong He,Maosen Guo,Minqiang Hu
DOI: https://doi.org/10.1109/tdei.2023.3289478
IF: 2.509
2023-01-01
IEEE Transactions on Dielectrics and Electrical Insulation
Abstract:Long-term exposure to high temperatures can lead to the physicochemical properties of DC cross-linked polyethylene (XLPE) cable material degrading irreversibly, compromising its electrical performance. In this work, the effects of thermal aging on electrical tree initiation and the performance of treated cable material were examined. A Dam-flood model was constructed to investigate the effect of thermal aging on electrical tree initiation behaviors, where indices corresponding to aging performance and electrical tree initiation behaviors were obtained based on this model. Ultimately, by means of multiple linear regression estimation and a GM (1, N) model, prediction methods for the electrical tree initiation voltage (U50%) and electrical tree initiation probability (P) were proposed. Results indicate that U50% and P do not exhibit a monotonic change with an increase in thermal aging time and, based on the Dam-flood model, the recrystallization reaction can be regarded as the primary cause for this phenomenon. According to the Dam-flood model, the indices for gel content (Cge), crystallinity (Xc), and dielectric DC breakdown strength (Eb) are significantly correlated with U50%, and these three indices can be used to determine the changing regularity of U50%. Furthermore, as the relative errors of the validation samples are all less than 15%, an equivalent relationship between P and the indices corresponding to thermal aging performance can be established. This could be used in future research to determine the electrical tree initiation of XLPE cable.
engineering, electrical & electronic,physics, applied