Research
Research Overview
My research focuses on the intersection of Power Systems, Energy, Control, and Machine Learning. I work at the Smart Grid Lab, Center for Energy Systems Research (CESR), Tennessee Technological University, Cookeville, TN, USA.
The overarching goal of my research is to develop intelligent, efficient, and resilient energy management systems for modern smart grids with high penetration of renewable energy sources and distributed energy storage systems.
Current Research Interests
Energy & Power Systems
- Energy Management Systems (EMS) — Developing optimal dispatch strategies for community microgrids
- Renewable Energy Integration — Solar PV, wind energy integration into smart grids
- Battery Energy Storage Systems (BESS) — State-of-charge balancing, peak shaving, grid-scale storage
- PV Hosting Capacity Analysis — Distribution network hosting capacity studies
- Peak Shaving Strategies — Demand response and load management
- Power System Operation & Planning — Economic load dispatch, optimal power flow
- Power Electronics Converters/Inverters — Dual Active Bridge converters for EV charging
- Power Quality & System Protection — Grid stability and protection schemes
Control & Simulation
- Adaptive Critic Design — Reinforcement learning-based adaptive control
- Controller Hardware-in-the-Loop (CHIL) Simulation — Real-time hardware-in-the-loop testing using Typhoon HIL
- Micro-grid & Smart-Grid Systems — Decentralized control and consensus algorithms
- Deep Reinforcement Learning for EMS — DQN, Meta-DQN, Safe-RL for energy management
- Metaheuristic Optimization — Swarm intelligence, evolutionary algorithms
AI & Machine Learning in Power Systems
- Deep Learning for Energy Forecasting — Hybrid deep learning models for solar irradiance forecasting
- Anomaly Detection — Smart grid anomaly detection using AI/ML techniques
- Cybersecurity in Smart Grids — Man-in-the-Middle attack detection, Modbus vulnerability analysis
- Digital Twin Applications — Digital twins for autonomous vehicular systems
Ongoing Projects
Meta-DQN-Tuned Safe Reinforcement Learning for Community Microgrid EMS
Developing a safe deep reinforcement learning-based energy management system for community microgrids that ensures constraint satisfaction while optimizing economic objectives. Uses a Meta-DQN architecture to tune the safe RL policy.
Intelligent Energy Management Systems using Controller Hardware-in-the-Loop with Deep Reinforcement Learning
Real-time implementation of intelligent energy management systems using controller hardware-in-the-loop (CHIL) simulation with deep reinforcement learning algorithms in Typhoon HIL-Python co-simulation environment. Optimizes microgrid dispatch, demand response, and battery management through advanced control techniques.
Quantum-Enhanced Recurrent Neural Networks with Transfer Learning for Solar Irradiance Forecasting
Hybrid quantum-classical approach combining quantum-enhanced recurrent neural networks with transfer learning techniques for accurate day-ahead solar irradiance forecasting, incorporating spatial-temporal information and meteorological constraints.
Software & Tools
| Category | Tools |
|---|---|
| Simulation | MATLAB & Simulink, Typhoon HIL, OpenDSS, PLECS, LTSpice |
| Programming | Python, C, C++, MATLAB |
| Documentation | LaTeX, MS Word, MS Excel, Visio, PowerPoint |
| Hardware | Typhoon HIL 606, RTAC, Waveforms (Analog Discovery 2) |
| RF/EM Simulation | ADS (Advanced Digital Design), Ansys Electromagnetics |
| Web | HTML, CSS, JavaScript, PHP, WordPress |
Research Lab
I conduct my research at the Smart Grid Lab, Center for Energy Systems Research (CESR), Tennessee Technological University.
Location: Clement Hall, Room 103, Cookeville, TN 38505
The HILLTOP (Hardware-in-the-Loop Laboratory Testbed and Open Platform) enables real-time controller hardware-in-the-loop testing with Typhoon HIL 606 devices, allowing realistic simulation of power electronic converters, battery storage systems, and communication protocols (Modbus TCP/IP, IEC 61850).