Privacy-Preserving Federated Learning for Value-Added Service Platform in Advanced Metering Infrastructure. / Zhang, Xiaoyu; Córdoba-Pachón, José-Rodrigo ; Watkins, Chris; Kuenzel, Stefanie.

In: IEEE Transactions on Computational Social Systems, 08.04.2022.

Research output: Contribution to journalArticlepeer-review

Submitted

Abstract

The advanced metering infrastructure (AMI) not only provides near real-time two-way communication between the consumers and the utility but also enables third parties to provide relevant value-added services to the consumers to improve user satisfaction. However, existing services are implemented in a centralized manner which has potential and associated privacy risks. To better balance the quality of the services and ensure users’ privacy, a third-party AMI service platform based on differentially private federated learning is proposed in this paper. In the proposed system, the neural network models are trained locally, and only model parameters are shared with the central server. Moreover, the identity of individuals is eliminated by adding random Gaussian noise during the secure aggregation. Furthermore, an attention-based bidirectional long short-term memory neural network model is adopted to solve the long-range dependency problem of conventional neural networks. In the case study, a residential short term load forecasting task is implemented to evaluate the performance of the proposed model. Compared with other state-of-the-art models, the proposed model can achieve similar accuracy as the typical centralized model and balances the trade-off between privacy loss and prediction accuracy flexibly. The proposed method can defend the system from both inner and external attackers efficiently.
Original languageEnglish
JournalIEEE Transactions on Computational Social Systems
Publication statusSubmitted - 8 Apr 2022

ID: 44289234