Ali Peivand | Searcharxiv

arXiv2026

Integrating Cybersecurity in Predictive Cost-Benefit Power Scheduling: A DeepStack Model with Dynamic Defense Mechanism

This paper introduces a novel, deep learning-based predictive model tailored to address wind curtailment in contemporary power systems, while enhancing cybersecurity measures through the implementation of a Dynamic Defense Mechanism (DDM). The augmented BiLSTM architecture facilitates accurate short-term predictions for wind power. In addition, a ConvGAN-driven step for stochastic scenario generation and a hierarchical, multi-stage optimization framework, which includes cases with and without Battery Energy Storage (BES), significantly minimizes operational costs. The inclusion of DDM strategically alters network reactances, thereby obfuscating the system's operational parameters to deter cyber threats. This robust solution not only integrates wind power more efficiently into power grids, leveraging BES potential to improve the economic efficiency of the system, but also boosting the cyber security of the system. Validation using the Illinois 200-bus system demonstrates the model's potential, achieving a 98% accuracy in forecasting and substantial cost reductions of over 3.8%. The results underscore the dual benefits of enhancing system reliability and security through advanced deep learning architectures and the strategic application of cybersecurity measures.

eess.SY↗

arXiv2026

Byzantine-Resilient Federated Multi-Agent Optimization Framework for Cyber-Secure Interconnected Microgrids

The escalating digitalization of distribution networks has exposed interconnected Microgrid (MG) clusters to Stealthy False Data Injection Attacks that bypass Bad Data Detectors and propagate through tie-line couplings and shared learning channels. This paper proposes BR-FedMAPPO, a Byzantine-Resilient Federated Multi-Agent Proximal Policy Optimization framework that learns a triple-surface Moving Target Defense and an adaptive isolation strategy for cyber-secure operation. Each MG hosts a local Actor-Critic Agent whose policy is partitioned into a globally federated shared encoder and a privately retained action head, so no MG exposes the configurations, cardinality, or locations of its D-FACTS lines, Battery Energy Storage (BES) units, or tie-line capacities. The action vector perturbs D-FACTS reactances, redirects BES injections, reshapes inter-MG exchanges, and includes a continuous islanding signal. A two-stage Byzantine-resilient aggregation rule combines trimmed-mean filtering with reward-weighted updates. This scheme incorporates a detection-quality score based on the F1-score and False Positive Rate to penalize clients causing false alarms. Simulation results on four interconnected MGs based on the IEEE 30- and 118-bus test systems demonstrate effective mitigation of coordinated S-FDI attacks, containment of cascading disruptions through adaptive isolation, and protection of distributed learning channels against malicious model manipulations while maintaining cost-aware dispatch performance.

eess.SY↗