High-fidelity, Large-scale, Realistic Dataset Development

Default ARPA-E Project Image


Program:
GRID DATA
Award:
$784,945
Location:
Atlanta,
Georgia
Status:
ACTIVE
Project Term:
05/15/2019 - 06/30/2023

Critical Need:

Several emerging issues, including the resiliency of electric power delivery during extreme weather events, expanding use of distributed generation, the rapid growth of renewable generation and the economic benefits of improved grid efficiency and flexibility, are challenging the way electricity is delivered. This grid of the future requires advances in transmission and distribution system management with algorithms to control and optimize how power is transmitted and distributed on the grid. However, the development of these systems has been hindered because the research community lacks high-fidelity, public, large-scale power system models that realistically represent current and evolving grid characteristics. Due to security and privacy concerns, much of the real data needed to test and validate new tools and techniques is restricted. To help drive additional innovation in the electric power industry, there is a need for grid models that mimic the characteristics of the actual grid, but do not disclose sensitive information.

Project Innovation + Advantages:

Georgia Tech will generate publicly releasable large-scale, high-fidelity datasets using techniques developed under GRID DATA funding (the team was originally funded as the University of Michigan). These datasets will be based on the RTE transmission system and conform to the technical and mathematical requirements of the Grid Optimization (GO) Competition’s Challenge 2, which focuses on the security-constrained optimal power flow (SCOPF) problem. SCOPF takes preventive and corrective scenarios into account. Georgia Tech will validate the feasibility and realism of these datasets to ensure they are high quality and as challenging as real transmission systems. The team will also convert datasets generated by other GRID DATA teams to conform to the Challenge 2 formulation. Both generated and converted datasets will be used to facilitate Challenge 2, after which they will be publicly released. These datasets will become vital assets to power systems researchers because they will enable publishable and reproducible research on a fast-evolving power grid. They will open the door to research on the wide-scale incorporation of intermittent resources (such as wind and solar), distributed resources, and storage. This research will enable reliable and economical grid planning and operation that optimally incorporates these innovative technologies, transforming the power grid.

Potential Impact:

If successful, Georgia Tech’s project will accelerate the development of new power system optimization algorithms by enabling more comprehensive and transparent testing.

Security:

New grid optimization algorithms could increase the grid’s resiliency and flexibility, improving its security during extreme weather and other threats.

Environment:

The team’s technology could enable greater integration of renewable electricity onto the grid, which would help reduce reliance on carbon-emitting, fossil fuel generation.

Economy:

The project could lead to greater efficiencies for grid operators and power generators and therefore help reduce operating costs.

Contact

ARPA-E Program Director:
Dr. Richard O'Neill
Project Contact:
Pascal VanHentenryck
Press and General Inquiries Email:
ARPA-E-Comms@hq.doe.gov
Project Contact Email:
pascal.vanhentenryck@isye.gatech.edu

Related Projects


Release Date:
01/15/2016