Implementation Of Multi-Method Fault Identification On 150 Kv Power Transformer

One of the vital and expensive pieces of equipment in an electric power system is transformer. It designed to have good reliability to work for years, expected to operate for more than 40 years. However, during its operational lifetime transformer condition is decreasing overtime. Dissolved Gas Analysis (DGA) is a diagnostic tool of transformer to assess the transformer to anticipate sudden failure. Recently, many interpretation methods were developed to interpret transformer fault based on DGA. Multi-method interpretation is the newest fault interpretation technique to identify transformer fault by combining another methods interpretation. Roger Ratio, IEC Ratio, Duval Triangle, and Duval Pentagon are the methods used in Multi-method by applying scoring and weighting into each interpretation result. The highest score from fault classifications will represent the result of Multi-method interpretation. Through comparison of all methods used, Multi-method achieve the best accuracy among them. It’s proven that the use of multi-method could increase interpretation accuracy and reliability. This research implements Multi-method interpretation on 3 UPT Malang fault indicated transformers, with fault prediction result of T1 (Thermal Fault below 300℃) in transformer A and B, and T2 (Thermal Fault between 300℃ to 700℃) in transformer C. Multi-method interface is also developed to help the fault interpretation process in UPT Malang transformer. The use of Multi-method interpretation could reduce error in fault identification that led to mishandling and unnecessary cost management.