Increasing Line Outage Localization Performance with Ensemble Classifiers
Machine Learning
Summary
The authors studied how to better identify which power line failed in an electricity grid by looking at data from other lines. They tested different machine learning methods and found that using a group of models working together (ensemble classifiers) worked better than just one model alone. They also showed that carefully choosing which lines to observe using a smart selection method (greedy MCP) improved their ability to pinpoint outages. One particular technique called extra-trees bagging gave the best results in many tests. Their results were backed up by statistical evidence.
Authors
Daniel Flores, Yuanrui Sang, Michael P. McGarry
Abstract
In many cases, the outage of one transmission line in a system can be localized by monitoring the power flow of another line, and machine learning methods can be used to distinguish the cases under uncertainty. In this study, we examine the improvements in line outage localization performance achieved by various ensemble classifiers compared to single-model methods. In the case studies, we compared the classification results with measurement data collected at observed transmission lines (OTLs) selected using three algorithms, i.e, greedy maximum coverage problem (MCP), high-eta, and random selection, based on two sensitivity factors, i.e., line outage distribution factors (LODFs) and line outage impact factors (LOIFs). We found that the OTLs selected by the greedy MCP algorithm yielded the highest F1 score and the ensemble classifiers significantly outperformed a base kNN classifier. The extra-trees bagging technique achieved the highest F1 score in many instances. All the findings were statistically significant.