To ensure a reliable operation of power system, power transformers’ health condition must be continuously monitored and assessed for appropriate operation and maintenance decisions. Currently, there are various insulation condition monitoring of power transformers. Such as dissolved gas analysis (DGA), Duval Triangular Method (DTM) and Rogers Ratio Method (RRM) Artificial neural network (ANN) etc. In the paper, 132/33 kV Bauchi substation was used as the study system and a selected injection feeder simulated for fault impact tests using Power System Computer Aided Design (PSCAD). The results which emanate from the study revealed that the current magnitude of bus voltage was 7.472%, bus voltage drop below 5% recommended while the oil and impregnated paper insulation degradation characterized with Kelman Transport X test equipment and MATLAB based ANN fault detection and classification jointly correlated that the sample oil only suffered for thermal faults greater 7000C but other characteristics such as arc faults, water contents were 73 ppm etc remained satisfactory for in the reference substation. Therefore this paper has demonstrated detection, location and restoration of faults in the installed transformers.
Power Transformer, Dissolved Gas Analysis, Fault Diagnosis Tools, Artificial neural network
Laouali M. Mamane , Abdullahi L. Amoo , Jibril D. Jiya "Artificial Neural Network Based Fault Diagnostic Tools of Power Transformer in Bauchi 132/33 kV Substation" Iconic Research And Engineering Journals Volume 5 Issue 5 2021 Page 146-154
Laouali M. Mamane , Abdullahi L. Amoo , Jibril D. Jiya "Artificial Neural Network Based Fault Diagnostic Tools of Power Transformer in Bauchi 132/33 kV Substation" Iconic Research And Engineering Journals, 5(5)