Identifying Transformer Incipient Events for Maintaining Distribution System
Reliability
Abstract
This paper presents the time domain and timefrequencydomain analysis results of incipient events insingle-phase distribution transformers. This analysis willaid in the development of an automatic detection methodfor internal incipient faults in the transformers. Thedetection method can provide information to predictfailures ahead of time so that the necessary correctiveactions are taken to prevent outages and reduce downtimes. The analyzed data was obtained from simulationsand experiments. Time-frequency analysis was performedusing Discrete Wavelet Transform (DWT). The obtainedresults are discussed
Keywords: Transformer, incipient fault, discrete wavelettransform
1. Introduction
The trend toward a deregulated global electricitymarket has put the electric utilities under severe stress toreduce operating costs, enhance the availability of thegeneration, transmission and distribution equipment andimprove the supply of power and service to customers.Further electric power systems was identified by the U.S.government as a critical infrastructure. Incipient faultdetection in transformers can provide information topredict failures ahead of time so that the necessarycorrective actions are taken to prevent outages and reducedown times. Hence protecting a critical infrastructure.The faults that occur within the transformer protectionzone are internal faults. Transformer internal faults can bedivided into two classifications: internal short circuitfaults and internal incipient faults. Internal short circuitfaults are generally turn-to-turn short circuits or turn toearth short circuits in transformer windings. These faultsoccur suddenly and usually require fast action byprotective devices to disconnect the transformer from theelectric system [1].Incipient transformer faults usually develop slowly,often in the form of a gradual deterioration of insulationdue to some cause. When the condition of systemequipment degrades because of some electrical, thermal orchemical effects, intermittent incipient faults begin topersist in the system. If not detected, this may lead tocatastrophic failure, which degrades quality of service andeventually longer outages. Service cannot be restoreduntil the source of the failure is repaired [2-3]. Majordamage may require shipping the unit to a manufacturingsite for extensive repair, which results in an extendedoutage period. If the condition can be detected beforemajor damage occurs, the needed repairs can often bemade more quickly and unit placed back into servicewithout prolonged outage.Incipient faults are not detectable at the transformerterminals using the normal protection methods such asfuse and relay protection. Over the years, variousincipient fault detection techniques, such as dissolved gasanalysis [4] and partial discharge analysis [5] have beensuccessfully applied to large power transformer faultdiagnosis. Since these techniques have high-cost andsome are off-line, a low-cost, on-line incipient faultdetection technique for distribution transformers usingterminal measurements is being proposed. Accordingly, amethod for on-line condition monitoring of distributiontransformers based on advanced signal analysis isdeveloped to detect internal incipient faults. Wavelets,which provide greater resolution in time for highfrequency components of a signal and greater resolutionin frequency for low frequency components of a signal,

Download full report
http://citeseerx.ist.psu.edu/viewdoc/dow...1&type=pdf