15 kVA, 11kV/400V Transformer Winding Fault Detection using Sweep Frequency Response Analysis (SFRA)

Abstract

Power transformers offer safety to electrical systems as they have a weakness that not only reduces system efficiency but also results in unscheduled outages and other sudden failures even after routine maintenance. The ageing of the insulation as well as short circuit issues, might be the primary causes of failure in the transformer winding. The problem is solved by creating the frequency response analysis (FRA) technique which improved in the diagnosis and detection of power transformer failures. By applying a sweep frequency to the winding, this technique detects winding deformation. The sweep frequency response analysis (SFRA) is simulated using MATLAB/SIMULINK software and the results is used to determine if the transformer is in unfault or fault condition by evaluating the frequency variations in the transformer winding. According to the analysis, when a fault is added into the winding, the first resonance is greater than the unfault conditions. From the unfault condition, it can be concluded that the first resonance after transformer model simulation occurs at 414 Hz, while for the inter turn fault condition, the first resonance of 415 Hz occurs at the second winding. Furthermore, for turn to turn fault condition is concluded that 415 Hz of first resonance occurred between first and second winding. Lastly, after changing the ground capacitance to 52.6 pF due to changing the length between windings to 0.0209 meters, it is recorded that the first resonance were occurred at 189 Hz. Based on this condition, the lower capacitance value reflects the life span of the transformer spent due to the lower frequency resonance.