No abstract

No abstract

This paper proposes data-mining-based models to diagnose outage data in distribution power systems. In this work, outage data from a local distribution company was gathered and aligned with weather data. Then, a subset of features was selected to reduce the processing time and simplifying purposes. To increase the fairness of the nal models and to account for di erences in misclassi cation cost, a customized cost matrix was used. Two decision-tree-based modeling algorithms were trained and tested. Results showed the ability of the established models to diagnose the root cause of an outage fairly well. In addition, an ensemble of the decision-tree-based models was built, which outperformed the other two models in almost all cases. Finally, applications of such models to decreasing outage duration and improving the reliability of the power distribution network were investigated.