Multi-core Identification of Mixed Power Disturbances

In order to solve the problem of edge overlap between typical classification features of mixed power disturbances, a new method that uses multiple feature extraction methods to identify the types of disturbances is proposed in this paper. Firstly, the different characteristic curves is analyzed to illustrate the effectiveness of multiple feature extraction methods in the identification of mixed disturbances. Secondly, in order to consider the correlation between high-dimensional features and target categories and the standardization of measurement scales, the improved maximum correlation minimum redundancy criterion is used to select the key feature subsets that are effective for identification, and then the multi-core SVM(Support Vector Machine) that takes into account the radius information is used to identify the mixed power disturbances. Finally, the simulation results show that the identification algorithm proposed in this paper can effectively identify various disturbances under different noise intensities, which demonstrates the effectiveness and feasibility of the proposed algorithm. This method overcomes the influence of mixed power disturbance feature space ambiguity on the identification accuracy, which is less affected by noise and has good stability.