Welcome to the website of the Smart Energy reseARCH (SEARCH) group, part of Johns Hopkins University's Ralph O'Connor Sustainable Energy Institute (ROSEI).
Our objective is to reinforce economic efficiency, improved electricity access and affordability, environmental responsibility, reliability, resiliency, and security of electric power and energy supply.
Evidence shows that trade-offs between these objectives are best resolved through markets, not command and control - regulation does not build power plants, interconnect new loads or move power across transmission lines, but it can be smart, adaptive and non-intrusive so that it can help markets do all of that better.
Our work is guided by our values.
Recent Announcements
- Data Center Integration + Policy: Shift or Curtail? How Much Data-Center Flexibility Is Worth Depends On The Host Power Grid
- Supply chains + Policy: Grid-Supporting Equipment Supply Chains Constrain the Feasible Pace of Power System Expansion
- Supply chains + Generation Expansion: Generation Expansion Planning With Upstream Supply Chain Constraints on Materials, Manufacturing, and Deployment
- ML for Energy (Theory): Reachability Guarantees for Energy Arbitrage
- ML for Energy (Application): Accelerating Underground Pumped Hydro Energy Storage Scheduling with Decision-Focused Learning
- AI and the Environment (Perspective): Metrics to Enable Flexibility of Non-Hyperscale AI Usage
- Participated in documentary on the AI/energy race between the US and China (300k+ views on Youtube alone)
- Commented in Newsweek on Texas's directive on data centers, electricity prices, and the limits of policy mandates
- Commented in CNBC and Fox Baltimore on White House's Execuitve Order to secure supply chains for grid-supporting equipment
- Commented in Politico on granularity of carbon accounting
- Owen Reed graduated with a M.Sc. degree and moved to Harvard for his Ph.D.
- Zhirui Liang started as an assistant professor at the University of Calgary.
- Anwar Khan moved from BSG to Solar Landscape as their Direct, Energy Markets Strategy.
Our Work
Electricity demand is growing again after two decades of near-zero growth, driven by data centers, electrification, and new industrial load. Meeting it is critical for US success, as a nation and the leader of the free world, and requires infrastructure decisions that are expensive, long-lived, and made under deep uncertainty. We apprecaite rigor and theory in our work, but we aim it to be useful for policymaking in Baltimore, Maryland, and beyond.
Current interests include:
- AI and optimization: decision-focused learning, inverse optimization, deep uncertainty
- Markets and resource adequacy: capacity accreditation, storage guarantees, extreme events
- Data centers and new loads: flexibility, interconnection, cost allocation, supply chains
- Transmission: multi-criteria expansion, grid enhancing technologies, offshore networks
- Security: grid-edge cybersecurity, cyber insurance and economics of cyber-resiliency
- Energy transition: industrial flexibility, new nuclear, carbon accounting
- Policy: helping various decision makers to make the right choices, in a economy-wide context
Methodologically, we focus on mathematical modeling of power grids together with interdependent infrastructure systems, and on control, optimization, and system-theoretic methods for their operation and planning. Two things make this hard: representing grid physics accurately requires NP-hard computation over large-scale, hierarchical, multi-agent networks, and growing customer-side and renewable resources bring uncertainty and limited controllability that obscure the value of these important technologies. Taken together, these two challenges push operators toward conservative practices that are costly and wasteful, often without being more reliable, and leave regulators approving suboptimal decisions on timelines that cannot keep pace with load growth. Our methodological work targets exactly that gap.
We gratefully acknowledge our project sponsors and partners:
