Implementation of Machine Learning in the Power Transformer Maintenance Prioritization Information System
Abstrak
Electricity plays an irreplaceable role in human daily life. To meet the increasing demand for electrical energy, a reliable electrical system, such as a power transformer, is required. Power transformers hold a crucial role in the electrical power system, where the long-term reliability of the transformer is closely related to the safety and stability of the power system. Therefore, transformer maintenance must be carried out to anticipate sudden failures and ensure the overall reliability of the electrical power system. It is essential to conduct health assessments to determine the condition of power transformers. These assessments can be performed in various ways, including the Health Index and Dissolved Gas Analysis. The Duval Pentagon Method and Duval Triangle Method are used in Dissolved Gas Analysis to ascertain the condition of transformers. In this development, a comparison of three machine learning models—Support Vector Machine, Artificial Neural Network, and Random Forest—was made using the Duval Pentagon Method and Duval Triangle Method datasets to obtain the model with the highest accuracy. The model with the highest accuracy will be implemented in a transformer maintenance information system to determine the transformer's condition. From the comparison results, it was found that the Random Forest model achieved the highest average accuracy of 99.3% on the DTM dataset. Meanwhile, on the DPM dataset, the model with the highest average accuracy was the Support Vector Machine (SVM) with a score of 97%. In this system, Health Index calculations with oil and paper as test result parameters are also applied. The results of the Health Index and Dissolved Gas Analysis calculations can determine the appropriate recommendations for transformer actions. Thus, the system can facilitate transformer experts in maintaining the health of power transformers.