Signal Processing Application in Fault Diagnosis of Three Phase Transformers

Data acquisition and subsequent fault diagnosis of a three-phase transformer is presented in this paper. A PC with data acquisition card is used for the purpose of storing, monitoring and diagnosis. Spectral analysis algorithms based on FFT as well as wavelet transform are used to obtain the signature of currents under various normal and abnormal conditions. The wavelet transform is implemented to extract the features from the transient signals simultaneously both in time and frequency domain. Current signatures so obtained serve as input for classifying various faults.