Attack detection and positioning algorithm based on improved decision tree support vector machine

In response to network attacks on the intelligent power grid automatic power generation control (AGC) system, a detection and positioning algorithm based on the improved decision tree support vector machine (DT-SVM) algorithm is proposed. This algorithm achieves the judgment of attack types through hierarchical combinations based on the improved SVM binary classifier, and arrange the classification roles of different levels in the decision tree based on the distance between categories and the detection accuracy of the binary classifier to avoid error accumulation. Simulation verification was carried out based on the IEEE 39-node model, and the simulation results show that the proposed improved DT-SVM algorithm has good detection performance.