Power Quality Disturbance Detection using Machine Learning Algorithm

The challenge of Power Quality Disturbances (PQDs) is now admitted as a crucial characteristic of a power system network. For structured power quality, disturbance causes must be recognized and regulated. This is accomplished through detection and classification of different PQDs. This article suggests a methodology to detect and classify PQDs using machine learning algorithm. The features of the signals are extracted through a mathematical morphology filter. These features are input to train one of the machine learning algorithm called Decision Tree (DT) and builds DT model to test and classify PQDs. For the purpose of classification, ten different types of disturbances are considered in this work. The proposed method is demonstrated on a data which is generated through PQD model using MATLAB. The performance of the proposed approach is good in PQD detection with accuracy rate of 99.95%