A QUANTUM- SECURED FEDERATED LEARNING FRAMEWORK FOR DYNAMIC SECURITY ASSESSMENT IN SMART GRIDS

The proposed Quantum Secured Federated Learning System for Smart Grid Dynamic Security Assessment (QFDSA) provides a secure and intelligent framework for monitoring and protecting modern smart grid infrastructures. By integrating Federated Learning with Quantum Key Distribution, the system enables collaborative model training across distributed smart grid nodes while ensuring strong data privacy and secure communication. The framework allows local nodes to train models using their own operational data without sharing sensitive information, thereby reducing privacy risks and communication overhead. The secure communication mechanism further protects the transmission of model updates from cyber threats such as eavesdropping or data manipulation.   The experimental evaluation demonstrates that the proposed system effectively performs dynamic security assessment by analyzing real-time smart grid data and detecting potential faults, instability, or cyberattacks. The use of federated learning improves the scalability and efficiency of the system, while the quantum-secured communication ensures the integrity and confidentiality of data exchange. As a result, the QFDSA framework enhances the reliability, resilience, and cybersecurity of smart grid operations.   In future work, the system can be extended by integrating advanced deep learning models to improve anomaly detection accuracy and predictive analysis. The framework can also incorporate edge computing techniques to further reduce latency and enable faster local decision-making at grid nodes. Additionally, the integration of blockchain-based security mechanisms may further strengthen trust and transparency in the federated learning process. Expanding the system to support large-scale smart grid networks and real-time energy management applications will further enhance its practical applicability in next-generation power systems.

1.                McMahan, B., Moore, E., Ramage, D., Hampson, S., & y Arcas, B. A. (2017). Communication-Efficient Learning of Deep Networks from Decentralized Data. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS). (Foundational paper on Federated Averaging - FedAvg). 2.                Kairouz, P., McMahan, H. B., Avent, B., et al. (2021). Advances and Open Problems in Federated Learning. Foundations and Trends® in Machine Learning, 14(1–2), 1-210. (Comprehensive survey of FL, including security and privacy aspects). 3.                Lo, H. K., Curty, M., & Tamaki, K. (2014). Measurement-Device-Independent Quantum Key Distribution. Nature Photonics, 8(8), 595-604. (Seminal paper detailing the MDI-QKD protocol). 4.                Pirandola, S., et al. (2020). Advances in Quantum Cryptography. Advances in Optics and Photonics, 12(4), 1012-1236. (Comprehensive review of QKD protocols and technologies). 5.                Xu, Y., et al. (2023). Adaptive resource allocation in Quantum Key Distribution (QKD) for federated learning. 2023 International Conference on Computing, Networking and Communications (ICNC). (Paper discussing QKD resource allocation for FL). 6.                P Kundur, J Paserba, V Ajjarapu, G Andersson, A Bose, C Canizares, N Hatziargyriou, D Hill, A Stankovic, C Taylor, T Van Cutsem, V Vittal. (2004). Definition and classification of power system stability IEEE/CIGRE joint task force on stability terms and definitions. IEEE Transactions on Power Systems, 19(3), 1387-1401. (Standard definitions for power system stability). 7.                E Vaahedi. (2014). Practical Power System Operation. John Wiley & Sons. (Textbook covering practical aspects of grid operation, including DSA). 8.                Liu, S., et al. (2021). Searching for the Devil in the Details: Learning Challenging Smart Grid Stability Boundary with Deep Learning. IEEE Transactions on Power Systems. (Example of deep learning for DSA). 9.                Sapountzoglou, N., et al. (2020). Charting the landscape of federated learning for healthcare. Frontiers in Digital Health. (Survey illustrating FL applications in sensitive domains). 10.             Bhagoji, A. N., et al. (2019). Analyzing federated learning through backdoor attacks. Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics (AISTATS). (Example paper on backdoor attacks against FL).