Diagnosis of transformer winding faults based on FEM simulation and on-site experiments

Abstract

Power transformers are the most expensive and important component in the power system, and may be subjected to breakdown due to mechanical and/or electrical faults. It is difficult to obtain the condition of the power transformer by only one test method, so transformer fault diagnosis based on different experimental and simulation methods is widely recommended. In this paper, the transformer winding faults diagnosis based on FEM simulation and on-site experiments is described. Based on the FEM simulation, it is necessary to analyze the effect of the winding faults on the change of the electrical parameters. The typical types of winding faults, such as axial displacement, and radial deformation, etc. are taken into consideration. Different fault degrees are investigated in the simulation analysis. Using the parameters calculated in the FEM model, an equivalent circuit of the power transformer is present in ATP/EMTP to study the performance of typical on-site experiments, such as frequency response analysis (FRA) and short-circuit impendence (SCI) test. The frequency bands, which might be treated as the diagnostic bands in the FRA, are simulated and compared. It is found that the resonances in the frequency bands of 200–250 kHz and 350–450 kHz are able to describe the situation, where the displacement percentage of the winding is varying from 0% to 9%. The SCI values under different fault severity are also obtained and compared. Therefore, the diagnosis method can directly bridge the types and degrees of the winding faults with the experiment results. The operating states of several 110 kV power transformers are tracked, and it can be found that the simulation results show a better consistency with the on-site experiment results, and the proposed diagnosis method will help to discriminate the winding faults.