An Intrusion Detection Method for Advanced Metering Infrastructure System Based on Federated Learning

An advanced metering infrastructure (AMI) system plays a key role in the smart grid (SG), but it is vulnerable to cyberattacks. Current detection methods for AMI cyberattacks mainly focus on data center or distributed independent node. On one hand, difficult train an excellent intrusion model self-learning other large amounts of are shared over network and uploaded central node training. These processes may compromise privacy, cause communication delay, incur high costs. With these limitations, we propose method based federated learning (FL). The deployed concentrators training, only its parameters communicated center. Furthermore, distributes each concentrator through aggregation weight assignments collaborative learning. optimized deep neural (DNN) exploited this proposed method, extensive experiments NSL-KDD dataset carried out. From results, improves performance reduces computation costs, delays, overheads while guaranteeing privacy.