Identifikasi Pendeteksian Arus Gangguan Hubung Singkat pada Sistem PT. PLN Dolok Sanggul Menggunakan Machine Learning

Abstract

In the electric power system, electrical system disturbances such as circuits can occur. Short circuits can be caused by many factors such as insulation failure due to excessive heat or humidity, mechanical damage to the equipment and failure to use the equipment. Electrical disturbances in transmission lines are an important problem in the electric power system, which has an impact on the stability and consistency of electricity distribution. Quickly identifying and classifying short circuit current disturbances is very important to overcome so that it is done rarely effectively to prevent greater system damage. Short circuit current disturbances can be detected using machine learning techniques and algorithms on PT's electrical network. PLN PLN Dolok Sanggul for machine learning requires current and voltage data parameters when a disturbance occurs and when there is no short circuit current disturbance. So this learning model can be divided into two parts, namely interference detection and classification. This machine learning training and classification is carried out in the Google Colaboratory application. By using various algorithm models for intrusion detection such as the SVM algorithm, decision trees, :KKN and random forests. So that machine learning accuracy is obtained which is evaluated for the most efficient algorithm, to be determined and used