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

Abstract

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

Federated Learning for Privacy-Preserving Intrusion Detection in Software-Defined Networks

“…Another research article by Zhao et al proposed [36] an IDS using FL for IoT environments. Liang et al proposed an IDS to detect cyber-attacks in advanced metering infrastructure (AMI) systems [37]. They used DNN as a training model with FL in their proposed model.…”

Advancing Power System Services With Privacy-Preserving Federated Learning Techniques: A Review

“…Data heterogeneity [4], [17], [18], [25], [27], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44] x x [45] x x [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60] x [61] x x x [28], [62], [63] x x [64] x x [7], [20] x x x x [8], [9], [19], [65], [66], [67], [68], [69], [70], [71],…”

“…Energy management systems [28], [37], [41], [46], [49], [58], [62], [78], [89] Preserving Anomaly detection [39], [43], [80], [87] High detection accuracy; improve the privacy; high communication efficiency; good scalability.…”

“…According to [130], many detection methods including statistical, classical analysis, and machine learning-based approaches) are applied to detect an anomaly. This subsection will focus on the machine learning techniques combined with FL to address communication efficiency issues and privacy concerns [39], [43], [80], [87]. Additionally, FL is employed to address the This article has been accepted for publication in IEEE Access.…”

“…The artificial metering infrastructure (AMI) system, crucial for information exchange between the power system devices, including energy management systems, DGs, and energy storage systems, is vulnerable to security threats. In [43], a federated method utilizing a deep neural network (DNN) model is developed for intrusion detection, and this method is implemented in the NSL-KDD dataset. This method outperforms centralized models in detection rate for R2L attacks and detection accuracy for U2R attacks, approximately 7% and 60% -70%, respectively.…”