Sensitive Information Identification Method of Power System Based on Deep Learning

Abstract

Power system is one of the most important infrastructures in modern society. In the power system, various sensitive information such as power supply status, load data and fault information need to be protected. In recent years, the methods based on deep learning has been widely used in the identification and protection of sensitive information in power systems. We propose a convolution neural network model based on pre-trained model and attention mechanism to classify and label power system data. Convolution neural network is a deep learning model, which offers a powerful and flexible tool for electric sensitive information detection. Pre-trained model and attention mechanism are two common technical means, which can improve the feature extraction and generalization ability of the model, thus providing effective support for image classification, target detection and other tasks. In the training process, the model we proposed realizes accurate and automatic recognition of sensitive information by learning the characteristics of input text information.