Innovative approach toward an algorithm for automated defect recognition for on-load-tap changers

Abstract

Power transformers are valuable assets in the power network, with the primary function to regulate the transmission and distribution voltage. The regulating device is the tap changer which sets the turn ratio. Tap changer failure is one of the leading causes of high-voltage power transformer failure. Most tap changer failures are caused by degrading contacts. Contact degradation may be the result of contact wear or of arcing-induced carbon deposition during on-load-tap (OLT) changes. Contact degradation increases contact resistance, causing increased heating and arcing, eventually leading to possible failure of the power transformer. Proper maintenance, therefore, requires a diagnostic system which can assess the condition of the contacts. The research presented is aiming at automated defect recognition and localisation from measured dynamic resistance measurement patterns, for two types of tap changers, the 'Diverter Switch' type and the 'Selector Switch' type OLT changer.