Notes from Open Energy Modeling Workshop – March 2020

Greg Schivley, US. PowerGenome – all the data for your US capacity expansion model — everything need to consider. Interested in adding modules to GenX to be able to expand it. The goal is to spend less time wrangling data, getting things in the right format, for the right regions, and to test sensitivities and policies. As everyone know, a large amount of data is needed (he lists).

PowerGenome is trying to provide all the data as well – 67 continental US regions – EPA’s IPM model, 2011 hourly demand profiles – but you can consolidate however you want. Have all the data on existing generators. Are going to develop a policy database and have all the data you need for future planning periods.

In order to manage multiple scenarios and periods. Help clean and open-source FERC 714 (hourly demand profiles). Has git-hub profile for extracting. Getting pretty close to a fully functional version that outputs for GenX, but want to do outputs for SWITCH and whatever other capacity expansion models people want to use. Also Jesse Jenkins, GenX. Bulk transmission constraints between 64 regions, at first.

Yixing Xu – validation if transmission capability between regions is matching…

Greg S. – currently counting on EPA’s model to be valid. Haven’t had chance to do validation of EPA model yet. They feel reasonably confident of it.

Consistency in terms of formats of data – Europe.

Building on puddle open source project to compile EIA data in the US, because we started from that, there was no attempt to have a consistent format with European projects. Would be happy to sit down with someone and do a comparison.

Taco Niet – Simon Fraser – hopefully gets into a machine readable format for easy conversion.

Greg – everything should end up in SQL-lite or Parquet databases easily readable into code, but not sure about other difficulties that might arise.

Daniel Olsen , Intellectual Ventures – US Test System with High Spatial and Temporal Resolution for Renewable Integration Studies

Trying to build the most high resolution open source dataset for the US electric grid to be used for production cost modeling. Running scenarios of high renewables future grids, to assess tradeoffs between different potential approaches for low carbon operation.

Summarizing scenario results to help inform policy makers

Dataset currently updated on zenodo – everything needed to support a multi-period optimal power flow – table format is MATPOWER compatible.

52 load zones, 13k generators, 82k buses, 104k branches, synchronized hourly profiles 8784 for all generators for 2016.

Generator cost curves, generator parameters, substation data, etc.

Paper on archive that talks about it in more detail

Grid topology, generator locations/capacities to match EIA form 860 (2016). Hourly demand by BAA from EIA disaggregated to load zones as necessary. Wind power from NOAA and solar from NREL. Hydro – EIA. Transmission capacities and cost curves.

Customized scenario options

Interconnections, renewable energy, transmission lines, demand, energy storage, scale-up in different areas.


Policy implications – CE expansion strategies, each state works to meet its own goals vs. collaborative. What see is a difference – can build fewer renewables or transmission. Effort is much higher to get to real decarb goals. Expanding analysis

Interactive online dashboard to visualize high level results. They are hiring.

More details on dataset develop:

Daniel Olsen, and Yixing Xu,