Comparison of Dissolved Gas Interpretation Techniques in Mineral Oil Immersed Transformers

One of the most important techniques for recognizing the potential fault in a power transformer at an early stage, i.e., incipient state, can be done through the interpretation of Dissolved Gas Analysis (DGA). Depending on the nature and seriousness of the fault, the insulation breakdown starts instantaneously, and the decomposition products will be distinct. DGA is periodically used to test the transformer's insulation oil to obtain dissolved gases that developed as an outcome of degradation in interior insulating materials. A conclusion can be drawn from the data obtained by DGA using distinct interpretation techniques. However, for the same instance, they could diagnose different fault categories. The purpose of this research is to examine the performance of different DGA interpretation techniques, i.e., Doernenburg Ratio Method (DRM), Rogers Ratio Method (RRM), IEC Ratio Method (IRM), Duval Triangle Method (DTM) and Duval Pentagon Method (DPM), [1] and evaluate these methods from fifty predetermined fault cases obtained from the IEEE Dataport in Enwen Li datasheet [2] using python. This study not only compares these techniques' overall accuracy but also focuses on the effectiveness of each method's ability to identify a particular fault separately.