Internal fault detection in transformers using wavelet analysis

This study represents a significant step towards the detection of internal faults in transformers. Initially, the normal operating conditions of the transformer, as well as phase-to-ground and phase-to-phase short circuit faults, were examined, which is a critical step in understanding the typical operation of the transformer. Subsequently, the current values obtained from the output were analyzed using Discrete Wavelet Transform through Matlab Wavelet Toolbox to detect faults. This method enables the detection of faults with high precision while also providing an efficient analysis process. As a result, the obtained wavelet transformation values were integrated into artificial intelligence models to enhance the fault detection process. This step automates the analysis process and enables faster and more accurate fault detection. This study can be considered as an important step towards the early detection and prevention of internal faults in transformers. The analyses conducted and the methods used can contribute to enhancing the reliability of energy systems, thereby contributing to a safer energy infrastructure. Keywords: Fault detection, internal fault, transformer, wavelet analysis, wavelet packet transform, wavelet, Matlab, differential protection, transient current.