Deep Learning-Based Control Model for Electronic Evidence Access Rights

With the rapid development of digitalization and intelligence in smart grids, how to identify untrusted users in the electric power system and implement user access control based on electronic evidence has become an important research topic. A credible assessment needs to be made on the subject users accessing electronic evidence. In this paper, we propose a deep learning-based user multi-dimensional attribute feature extraction access control model, which considers multi-dimensional attribute features such as user identity, behaviour, and device information, and determines whether the user is a legitimate access user by using CNN-Transformer deep learning framework to extract user features and determine their trust value. Through simulation experiments, it is proved that the access permission control model proposed in this paper has high accuracy and reliability and can meet the access requirements of different electronic evidence users.