In this research, a differential protection technique for a power transformer is proposed by using random forest and boosting\nlearning machines. The proposed learning machines aim to provide a protection expert system that distinguishes between\ndifferent transformer status which are normal, inrush, overexcitation, CT saturation, or internal fault. Data for 20 different\ntransformers with 5 operating cases are used in this research. The utilized randomforest and boosting techniques are trained using\nthese data. Meanwhile, the proposed models are validated by other measures such as out-of-bag error and confusion matrix. In\naddition, variable importance analysis that shows signalâ??s component importance inside a transformer at different instances is\nprovided. According to the result, the proposed randomforestmodel successfully identifies all of the current cases (100% accuracy\nfor the conducted experiment). Meanwhile, it is found that it is less accurate as a conditionalmonitoring element with accuracy in\nthe range of 97%â??98%. On the other hand, the proposed boostingmodel identifies all of the currents for both cases (100% accuracy\nfor the conducted experiment). In addition to that, a comparison is conducted between the proposed models and other AI-based\nmodels. Based on this comparison, the proposed boosting model is the simplest and the most accurate model as compared to\nother models.
Loading....