In the process of building a new audit cloud model,making full use of intelligent means to effectively detect network security vulnerabilities of audit business platforms and ensuring network and information security is an important measure for the power grid audit information revolution.Aiming at the current network attack detection problem of power grid audit busi-ness,this paper proposes an attack detection strategy based on deep convolutional neural network.The attack data are prepro-cessed,features are extracted by applying word frequency,and features are selected using an edit distance algorithm.This pa-per introduces the concept of fractional calculus into the earthworm algorithm,uses the improved fractional earthworm algo-rithm to train the deep convolutional neural network,and performs network attack detection and classification based on the se-lected features.It is verified by experiments that the method proposed in this paper can effectively detect network attacks and has a high accuracy.
维普
万方数据