<![CDATA[This study discusses the implementation of the Decision Tree algorithm for classifying electrical fault types based on data obtained from the Disturbance Fault Recorder (DFR) on a 150 kV transmission system. Disturbances in the electric power system can cause power instability and equipment damage if not detected quickly. The data used consists of 12 fault data including the fault current (I Fault), voltage during fault (V Fault), impedance (RΩ), and the cause of the fault. The classification process is carried out to distinguish the types of 1-phase, 2-phase, and 3-phase faults using a supervised learning approach with an entropy-based Decision Tree algorithm as a separation criterion. Model performance evaluation is carried out using the Leave-One-Out Cross Validation (LOOCV) method to maximize the use of small datasets. The test results show an accuracy of 91.67%, with the V Fault and I Fault parameters as the most influential features in the classification process. This study shows that the Decision Tree algorithm is capable of being an effective artificial intelligence-based solution for detecting electrical fault types quickly and interpretively. In the future, research can be developed by increasing the amount of data and conducting comparisons with other algorithms such as Random Forest and Support Vector Machine (SVM) to obtain more optimal results.]]>
Copyrights © 2025
