Power grid middle platform data anomaly detection based on two-stream convolutional neural network

With the rapid development of the Internet, network anomaly detection has increasingly become a problem worthy of attention. To effectively detect anomalies in the network, this paper describes a two-stream convolutional neural network (TS-CNN) anomaly data detection method. First, TS-CNN uses multiple 1-D convolutional layers to extract features from 1-D network data, and generate 1-D feature vectors through pooling; at the same time, TS-CNN preprocesses 1-D network data to generate 2-D grayscale images. Multiple 2-D convolutional layers are used to extract features from grayscale images, and 1-D feature vectors are generated through a pooling layer and a fully connected layer; finally, the two feature vectors are spliced through a fully connected layer to output the detection results. Experimental results show that the detection method based on TS-CNN can effectively detect network abnormal data.