Detection of Power Data Tampering Attack based on Gradient Boosting Decision Tree

Abstract

In order to improve the detection accuracy of the data tampering attack of the Power Cyber-physical System, a method of Power CPS attack detection based on Gradient Boosting Decision Tree is proposed in this paper from the angle of artificial intelligence. Firstly, by setting up the iForest anomaly value equation of physical system and IDS Shannon entropy function of information system, the feature extraction of data tamper detection is realized. Then, the CART decision tree is taken as the base learner of the detection classification, the target is minimized by the loss function, and the high intensity attack detection model is designed by iterative combination. Finally, the three-dimensional adaptive chaotic FOA algorithm is proposed for dynamic optimization of model parameters, an attack detection model under training optimal parameters. The analysis of the example shows that show that the proposed method can effectively detect power CPS data tampering attacks and has excellent detection precision.