The invention discloses a
power transmission system fault classification method based on a GA and an XGBoost-RF stacking
algorithm, and the method comprises the steps: obtaining an XGBoost-RF stacking
algorithm model through employing a limit
gradient boosting algorithm XGBoost as a base learning device, employing a
random forest algorithm RF as a meta learning device, and combining the two algorithm models through employing a stacking method in integrated learning, thereby achieving the fault classification of a
power transmission system. The performance limitation of the algorithm is successfully broken through, so that the faults can be classified more accurately; in order to solve the problem that the calculation time of integrated learning is slow, a
genetic algorithm is additionally added in the training process of a base learner for
feature selection, that is, the accuracy function of a limit
gradient boosting algorithm XGBoost is used as a
fitness function; according to the method, a binary coding
genetic algorithm (GA) is adopted to perform
feature data selection on original
feature data in a fault
feature data set, an optimized feature subset is obtained, the selected optimized feature subset has very good performance on a base learner and a meta learner, and due to the reduction of the number of features, the optimal feature subset is obtained. Therefore, the calculation time of the whole algorithm model is reduced while the accuracy is improved, and the whole stacking algorithm model is balanced in time and accuracy.