Weighted Islanding Detection for DC Microgrid Based on Random Forest Classification

At present, the main form of microgrid is AC grid. DC microgrids have received extensive attention and research with rapid development various power. The operation mode divided into grid-connected islanding operation. Islanding formed after circuit breaker tripped, which connects to large can be planned unplanned islanding. Unplanned will cause certain harm users systems, so it necessary detect accurately in microgrid. This paper proposes an detection method for based on random forest classification. Firstly, raw data cleaned, extracted features generated feature vector set. include six characteristic indexes, consist voltage, current, output active power their first order backward difference bus side. Then, classification, building model. model distinguish event successfully accurately. Based weighted more compared decision tree classification when processing amounts data.