Out-of-Step Protection in Generator Based on Intelligent Identification of Impedance Trajectory

Abstract

In order to improve the selectivity and quickness of generator out-of-step protection, a method of generator out-of-step protection based on support vector machine (SVM) for intelligent trajectory recognition is presented. Firstly, the motion feature is extracted from the measured impedance trajectory at the generator terminal, and the extracted feature sequence is calculated with statistical parameters to form 140-dimensional features. Secondly, the principal component analysis method is used to reduce the dimension of the feature to form the corresponding training input feature space, and the particle swarm algorithm is used to optimize the parameters of SVM. Finally, the simulation samples verify that the method can accurately identify the out-of-step oscillations. Compared with traditional out-of-step protection, this method improves the reliability and speed of generator out-of-step protection.