Skip to main content

Research portfolio

Research

Physics-grounded intelligence for IBR-dominant Power System

My research connects fast power-system dynamics with control, optimization, and trustworthy AI. The goal is to make modern grids with inverter-based resources secure, resilient, flexible, and economically efficient.

  • 2 research visions
  • 8 connected themes
  • Representative research outputs

Two complementary research visions

One vision develops the physical and mathematical foundations of IBR-dominant grids. The other develops intelligent models, agents, and infrastructure that operate within those foundations.

How the research connects

Across both visions, the portfolio follows a common path from representation to measurable engineering outcomes. AI appears throughout this path as both a method and an infrastructure.

  1. Models

    Represent physical and learned system dynamics at the fidelity decisions require.

  2. Boundaries

    Characterize stability, feasibility, capability, uncertainty, and safety limits.

  3. Decisions

    Embed those limits in control, operation, planning, markets, and engineering workflows.

  4. Outcomes

    Evaluate security, resilience, cost, reproducibility, and verifiability.

Research themes

Each theme states the core research questions and methods, highlights one representative work, and groups all related research outputs.

Vision · IBR-Dominant Power Systems

IBR Modeling, Control, and Stability

Develop decision-compatible dynamic models, stability certificates, and coordinated or adaptive control methods for heterogeneous IBRs and hybrid AC/DC systems.

  • Dynamic Modeling
  • Stability Certification
  • Grid-Forming Control
  • Adaptive Control

Representative work

Systematic Controller Design for Inverter-Based Microgrids With Certified Large-Signal Stability and Domain of Attraction

2023 · IEEE Transactions on Smart Grid

Explore questions and related outputs: IBR Dynamics, Control & Stability

Core research questions

  • Decision-compatible dynamic representation
  • Stability characterization and certification
  • Coordinated and adaptive IBR control

Related research outputs (18)

Vision · IBR-Dominant Power Systems

Network Operating Boundaries and Resource Capability

Characterize the geometry, computation, and limiting mechanisms of feasible regions, security regions, and network capability curves, and translate them into planning and operating constraints.

  • Operating Boundaries
  • Security Regions
  • Capability Curves
  • Boundary-Aware Decisions

Representative work

Operational boundary of flow network

2023 · Reliability Engineering & System Safety, vol. 231

Explore questions and related outputs: Operating Boundaries & Capability

Core research questions

  • Geometry and certification of feasible regions
  • Network capability curves and limiting mechanisms
  • Boundary-aware planning and operation

Related research outputs (24)

Vision · IBR-Dominant Power Systems

Dynamic-Security-Constrained Decisions

Embed fast dynamics, stability constraints, and controllable device capabilities into operation, scheduling, planning, co-design, and market decisions.

  • Dynamic Security
  • Stability-Constrained Optimization
  • Virtual Inertia Scheduling
  • Market Design

Representative work

Virtual Inertia Scheduling (VIS) for Real-Time Economic Dispatch of IBR-Penetrated Power Systems

2023 · IEEE Transactions on Sustainable Energy

Explore questions and related outputs: Dynamic-Security Decisions

Core research questions

  • Bridging fast dynamics and system-level decisions
  • Dynamic-security-constrained operation and planning
  • Co-design across assets, controls, and operations
  • Economic value and market representation of stability
  • Economic flows and dynamic market design

Related research outputs (15)

Vision · IBR-Dominant Power Systems

Resilience and Cyber-Physical Security

Quantify system resilience through reproducible event evidence and develop cyber-defense, extreme-event operation, and restoration methods tied to physical outcomes.

  • Resilience Metrics
  • Cyber-Physical Security
  • Extreme Events
  • Restoration

Representative work

A Review of Energy Storage for Power System Resilience: Functions, Metrics, and Applications

2026 · Applied Energy

Explore questions and related outputs: Resilience & Cyber-Physical Security

Core research questions

  • Resilience measurement grounded in event evidence
  • Cyber-physical monitoring and defense
  • Extreme-event operation and restoration

Related research outputs (8)

Vision · AI-Driven Intelligent Power Systems

Physics-Informed and Trustworthy AI for Power Systems

Embed physical laws, network structure, uncertainty, operational constraints, and safety certificates into learning models for safety-critical power-system modeling, control, and optimization.

  • Physics-Informed Learning
  • Safe Reinforcement Learning
  • Verifiable AI
  • Learned Dynamics

Representative work

Inverter PQ Control With Trajectory Tracking Capability for Microgrids Based on Physics-Informed Reinforcement Learning

2023 · IEEE Transactions on Smart Grid, vol. 15, no. 1

Explore questions and related outputs: Physics-Informed & Trustworthy AI

Core research questions

  • Physics-grounded and certified learning
  • Safe adaptation and generalization
  • Learnable dynamic representations

Related research outputs (7)

Vision · AI-Driven Intelligent Power Systems

Intelligent Engineering Agents and Automated Research

Develop domain-grounded LLMs and agents that turn engineering intent into executable, traceable workflows evaluated through physical and numerical outcomes under human supervision.

  • Large Language Models
  • Engineering Agents
  • Outcome Verification
  • Automated Science

Representative work

PFAgent: A Tractable and Self-Evolving Power-Flow Agent for Interactive Grid Analysis

2026 · arXiv preprint

Explore questions and related outputs: Engineering Agents & Automated Research

Core research questions

  • Executable engineering agents
  • Outcome-based agent verification and benchmarks
  • Human-supervised automated science

Related research outputs (3)

Vision · AI-Driven Intelligent Power Systems

Decision Intelligence for Operation, Planning, and Control

Connect forecasts, learned representations, and learning-enabled control to feasible, reliable, and computationally tractable engineering decisions under limited data, structural change, and uncertainty.

  • Forecasting
  • Uncertainty
  • Prediction-to-Decision
  • Learning-Enabled Control

Representative work

Fusion of Microgrid Control With Model-Free Reinforcement Learning: Review and Vision

2022 · IEEE Transactions on Smart Grid, vol. 14, no. 4

Explore questions and related outputs: Decision Intelligence

Core research questions

  • Forecasting under novelty and limited data
  • Prediction-to-decision coupling

Related research outputs (6)

Vision · AI-Driven Intelligent Power Systems

AI Infrastructure and Grid Co-Design

Co-design AI workloads and data centers with thermal, electrical, and grid layers to preserve computing service while improving reliability, economic performance, and environmental outcomes.

  • Data Centers
  • Flexible AI Workloads
  • Grid Interaction
  • Infrastructure Co-Design

Representative work

PowerMorph: Shaping LLM Training for Data Center Demand Response

2026 · IEEE International Parallel and Distributed Processing Symposium (IPDPS)

Explore questions and related outputs: Grid–AI Infrastructure Co-Design

Core research questions

  • Grid-compatible and energy-efficient AI workloads
  • Energy-system co-design for AI infrastructure
  • Lifecycle and environmental impacts

Related research outputs (3)

Explore the evidence and collaborations

The portfolio is a structured view of my research agenda. Publications document the evidence, while Projects describe sponsored programs and research roles.

View all publications Explore research projects Discuss collaboration