MADDPG-Based Security Situational Awareness for Smart Grid with Intelligent Edge

Advanced communication and information technologies enable smart grids to be more intelligent automated, although many security issues are emerging. Security situational awareness (SSA) has been envisioned as a potential approach provide safe services for power systems’ operation. However, in the cloud master station mode, massive heterogeneous terminals make SSA complicated, failure cannot promptly delivered. Moreover, dynamic continuous space also increases challenges of SSA. By taking advantages edge intelligence, this paper introduces computing between address drawbacks traditional paradigm. deep reinforcement learning algorithm based on paradigm multiagent deterministic policy gradient (MADDPG) is proposed. The minimum processing cost under premise detection error rate taken analyze grids’ Performance evaluations show that can achieve faster convergence optimal goal, namely provision real-time protection grids.