The transformer is the most important part of the electric power system, therefore maintenance needs to be carried out to prevent the emergence of anomalies in the transformer. Dissolved Gas Analysis (DGA) is a method for detecting anomalies in transformers. DGA is used to test the condition of the insulating oil in transformers by taking samples of the insulating oil. If an anomalous event occurs in a transformer, the resulting gas concentration will vary depending on the type of anomalous event in the transformer. The main problem underlying this research is the inability of previously existing anomaly detection systems to provide an optimal level of accuracy, traditional methods or approaches used also face obstacles in interpreting complex data from dissolved gas analysis. The aim of the research carried out is to be able to design an anomaly detection system on Transformers using DGA and to see the level of accuracy of the existing DGA method using KNN. In this research, the anomaly detection system on the transformer resulted in the highest level of accuracy being 94% using the key gas method and the lowest level of accuracy being 79% using the Doernenburg Ratio method. The conclusion of this research is that it is able to create a system that can make it easier to analyze anomalies in transformers, and can be used as an alternative method for determining the condition of transformers.
Copyrights © 2024