Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Grid Planning Tools topic
No spam. Unsubscribe anytime.
Sandia engineer Cody Noland outlines open‑source Quest tools for coordinated grid planning, flags large‑load modeling gaps
Summary
Cody Noland of Sandia National Laboratories described the open‑source Quest platform—capacity expansion 'Quest Planning', probabilistic 'Progress' and a production cost model—and said preliminary case studies show that large‑load flexibility and commissioning timing materially change storage and generation investment choices. He noted pilots with utility PNM and ongoing work to add stochastic large‑load behavior and AI workflow agents.
Cody Noland, an electrical engineer at Sandia National Laboratories, presented the Quest open‑source platform and its three core grid‑planning tools, saying Quest has been under development "for about 7 or 8 years now" and is intended to connect capacity expansion, probabilistic adequacy and high‑fidelity production‑cost analyses.
Noland said Quest Planning performs long‑term capacity‑expansion optimization that can vary spatial scope (from copper‑sheet to nodal) and temporal fidelity (seasonal blocks up to full 8,760‑hour analyses), while optimizing both power and energy (duration) capacities for a range of storage technologies. "We do allow for evaluation of a broad range of energy storage technologies… and we're also optimizing power and energy capacities," he said.
For probabilistic adequacy, Noland described the Progress tool as a Monte‑Carlo resource adequacy engine that captures weather and component failures to estimate outage frequency, duration and magnitude and includes battery degradation features. He said the Quest production cost model interfaces storage formulations with unit‑commitment and dispatch algorithms to analyze operational impacts under different market structures.
To manage inputs, model differences and outputs, Noland presented the Quest workspace, an interactive Python‑based interface for assembling and debugging workflows and interpreting results. He also detailed a Quest AI agent designed to map user prompts to model 'skills', assemble workflows in the workspace, run the tools and provide diagnostic summaries. Noland emphasized the agent "is not intended to replace planners" and should assist subject‑matter experts by surfacing data or modeling errors and summarizing results.
Pivoting to large‑load modeling, Noland said there are key gaps in how large loads (for example, data centers and other rapid growth loads) are treated across planning, adequacy and operations models. He described a proof‑of‑concept study on the RTS GMLC test system using synthetic large‑load profiles generated from a Markov process. That case compared scenarios that varied commissioning timing, configuration (grid‑supplied versus co‑located generation and storage) and flexibility allowances.
"The flexibility constraints really impacted the investments," Noland said, summarizing the case‑study finding that both flexibility and the timing of large‑load commissioning materially affected optimal generation and storage expansion and influenced the optimal duration of storage assets. He added that the most pronounced system‑level cost impacts in their runs showed up in fuel costs and that additional work is needed on workload‑shifting constraints and operational flexibility assumptions.
During Q&A, Rick Benitez of BNGC asked whether Quest has been used for interconnection studies at the RTO/ISO scale. Noland replied that Sandia has partnered with the utility PNM during early stages and has used synthetic test cases, but has not yet deployed Quest at an ISO/RTO level; he invited others interested in collaborations to contact Sandia.
Noland closed by acknowledging funding from the Department of Energy Office of Electricity Energy Storage Division and reiterated that Quest aims to be a transparent, extensible research framework that planners and researchers can integrate into their workflows.
The session moved to a break following the Q&A; presenters were thanked and the meeting was paused.
