Study on the Diagnostic Method of Fault Current and Excitation inrush Current

Aiming at the problems of poor classification effect under small sample problem and inefficient identification of massive monitoring data in the traditional fault diagnosis method of synchronous compensator, this paper proposes a novel online diagnosis method of regulator fault current and excitation inrush current. The method utilizes the acquired real-time data of three-phase current to construct a training set, and determines whether the input current is fault current through variable prediction model classification and training set. The simulation analysis shows that the recognition accuracy of the method is higher than other methods, and it has better adaptability to high-dimensional feature vectors.