Constructing a Security Monitoring and Protection Mechanism for Power Grid Data Throughout its Life Cycle based on Graph Convolutional Networks

Constructing a Security Monitoring and Protection Mechanism for Power Grid Data Throughout its Life Cycle based on Graph Convolutional Networks

연구 분야: Infrastructure

학회: 2025 3rd International Conference on Data Science and Information System (ICDSIS)

With the rapid advancement of smart grids, ensuring the stable operation of the power system hinges on the full life cycle security monitoring and protection of power grid data. However, traditional approaches, characterized by heavy reliance on manual experience and isolated system architectures, struggle to cope with the increasingly intricate and dynamic risks of network attacks and data leakage. This paper introduces a graph convolutional network (GCN)-based full life cycle security monitoring and protection mechanism for power grid data. Through experimental testing, it is concluded that the protection mechanism based on GCN generally attains accuracy values between 0.6 and 0.9, significantly surpassing the 0.2 to 0.7 range of traditional protection mechanisms. This method demonstrates the capability to accurately identify abnormal behavior in power grid data and effectively resist adversarial attacks, offering a promising solution for enhancing the security and reliability of smart grids.