My research develops trustworthy AI for cyber-physical systems, especially smart grids and critical energy infrastructure. It is organized around five themes. Each lists representative published work.
1. Cyber-Physical Security and Intrusion Detection
Focus: graph neural networks, physics-informed learning, attack graphs, adversarial detection and risk-based security for power systems.
- FADEG investigates feature-adaptive graph neural network ensembles for intrusion detection in cyber-physical power systems. IEEE Open Journal of Industry Applications, 2026
- A constraint-aware physics-informed neural network for intrusion detection in cyber-physical smart grids. Sustainable Energy, Grids and Networks, 2026.
- Topology-aware hierarchical attack graph optimization for cyber-physical power systems. Electronics, 2026
- Maximum entropy inverse reinforcement learning for cyber attack detection in IoT networks. IEEE Access, 2026
2. Trustworthy AI and Safe Reinforcement Learning
Focus: robust and resilient decision-making for grid operation under disturbances and adversarial attacks.
- SPARQ investigates soft Q-learning for cyber-resilient voltage regulation and autonomous grid operation. IEEE Access, 2025
- Adaptive PPO with multi-armed bandit clipping and meta-control for robust power grid operation under adversarial attacks. IEEE Access, 2025
- A review of deep reinforcement learning for power system stability control. IEEE Access, 2023
3. Smart-Grid Stability and Intelligent Control
Focus: transient stability assessment, voltage regulation and learning-based control during grid disturbances.
- Fast transient stability assessment using optimized temporal convolutional networks, recipient of the IEEE Industry Applications Society Second Prize Paper Award. IEEE Open Journal of Industry Applications, 2024
- A review of machine and deep learning for coherent power grid partitioning. IEEE Open Access Journal of Power and Energy, 2025
4. Federated and Distributed Intelligence
Focus: privacy-preserving learning, concept drift and distributed monitoring of power systems.
- FLACON applies deep federated transfer learning to transient stability assessment during symmetrical and asymmetrical grid faults. IEEE Open Journal of Industry Applications, 2024
- Online transient stability assessment under concept drift using adaptive random forests and federated learning for data streams. IEEE Internet of Things Journal, 2025
5. AI for Sustainable Infrastructure and Education
Focus: resilient energy systems for data centers, distributed energy resources, and AI-powered learning tools.
- An efficient framework for multi-DER integration in distribution networks with time-series demand flexibility. IEEE Access, 2025
- An adaptive learning platform for AI education with confidence-aware review and ML coding quests. IEEE ICELIE, 2026.
- Research software for data-quality assessment: see Software & Data.
Funded project: ReCIPeS-DC
Resilient Cyber-Physical Infrastructure for Sustainable Data Centers in Texas
| Role | Lead Principal Investigator |
|---|---|
| Funding | TEES Annual Research Conference (TARC) seed funding, Texas A&M Engineering Experiment Station |
| Award | 2026, $5,000 |
| Scope | Multi-institutional, AI-enabled cyber-physical resilience framework for data-center-integrated power systems in Texas |
The project aims to build AI-based methods that help power systems serving data centers detect and withstand cyber-physical disruptions. It builds on my work in smart-grid cybersecurity and resilient energy infrastructure. The project is at an early stage, and outputs will be added here as they are published.