In a two-part webinar presented by the USAID-NREL Partnership and the G-PST Consortium’s Open Tools and Data Pillar, expert speakers introduced open-source data platforms and tools that support planning for grid integration of variable renewable energy. The speakers discussed production cost modeling (PCM) tools for power sector planning and demonstrate PCM development, supported by an open-source data and code package for side-by-side learning. Tools covered included the Sienna applications, the System Advisor Model (SAM), and the Renewable Energy (RE) Data Explorer.

These open-source power system simulation tools as well as the world-class solar and wind resource data from RE Data Explorer are useful for stakeholders looking to evaluate the operational constraints and costs associated with different energy resource portfolios.

Speakers included:

  • Karin Wadsack, NREL
  • Kate Doubleday, NREL
  • Galen Maclaurin, NREL

This webinar was moderated by Holly Darrow, NREL.

Please navigate the box below to view the recordings of both sessions, code packages and resources for side-by-side learning, and frequently asked questions:

Following along with the webinar is optional. If you wish to do so, use the following resources to download the necessary materials and programs.

The code for the case study production cost modeling demonstration is available on Github: https://github.com/NREL-Sienna/PSI-Cambodia

An explanation of the demonstration is available in the PSI-Cambodia README, including some of the installation instructions below.

Instructions for installing software for the three open-source tools we will be using:

  1. RE-Data Explorer: Can be accessed online without any installation
  2. System Advisor Model (SAM): We will run SAM in Python using the PySAM wrapper, as well as designing plant specifications in the SAM GUI:
    1. Download the SAM GUI
    2. Install Python
    3. Activate the environment as described in the PSI-Cambodia README (step 2.c in the README)
  3. Sienna: Sienna is written in the Julia programming language
    1. Install Julia
    2. Activate the environment and run literate.jl as described in the PSI-Cambodia README (step 3 in the README)

For more information, visit Variable Renewable Energy Grid Integration Studies: A Guidebook for Practitioners.

View the presentation deck.

View the presentation deck.

View some of the frequently asked questions from this training below. For more questions from the Q&A portion of the event, view the event recordings.

If you have further questions, please reach out to USAID.NREL@nrel.gov or globalpst@nrel.gov.

The Renewable Energy (RE) Data Explorer:

  • Is it possible to download resource data from RE Data Explorer programmatically so things can be automated? Is there an API?
  • What is the source of RE Data Explorer’s transmission and power plant data?
    • The source of data is country specific. When you view a data set in RE Data Explorer, you can view the source by scrolling down to “Additional Information”.
  • What types of data are available on RE Data Explorer outside of wind and solar data?
  • Can RE Data Explorer be used with Python?
    • No, but you can access the solar and wind data via Python (see FAQ on APIs for resource data).

The System Advisor Model (SAM):

  • Can you simulate multiple generators in SAM?
    • SAM is a power plant-level design tool. SAM is not a system-level modeling tool and cannot be used to model the interactions between multiple generators connected at different points in the power grid. It does include some interactions that occur within an individual power plant, such as a wind farm with multiple individual turbines, and some hybrid plant models.
  • Can SAM account for a mix of renewable energy generation sources?
    • SAM is a power plant-level design tool, used to model one power plant at a time. It has a few hybrid plant models with interactions among technologies, including PV-storage or fuel cell-PV-storage plants. It is not used to model different generation “mixes” at the power system level; for that, please consider a capacity expansion and/or production cost model.
  • Can SAM account for different solar plant capacity sizes in the same model (without iterating the same plant at different locations)?

The Sienna Applications:

  • Do I need a high performance computer to run Sienna or is it available for use on Windows and/or MACOSX?
    • No, Sienna can be run on a standard laptop and can be used on both Windows and MACOSX. It can also be used on a high-performance computer or cloud computing system such as AWS.
  • Is there a SiennaOps tutorial and/or GUI?
  • Can I incorporate external transmission or powerplant data into Sienna?
  • Can Sienna represent battery operation?
    • Yes, SiennaOps includes a GenericBattery formulation that can be used to model utility-scale batteries and other types of energy storage.

Production Cost Models (PCMs):

  • Is it possible to account for a timeseries of demand on the grid when modeling power grids (for example, AC use when it is hot)?
    • Yes, a time-series of demand is one of the key inputs to a production cost model, which is used to determine the generation supply needed to meet that demand at each time step. In many power systems, the highest peak in that demand time-0series occurs in the hot months due to AC cooling demand; a production cost model can be used to verify supply will be able to meet that peak.
  • Can this type of modeling be done for smaller systems such as residential or off-grid systems?
    • Production cost modeling is best suited for transmission-scale or bulk power system modeling. For residential, off-grid, or microgrid systems, please consider using SAM for standalone PV or PV-storage systems or a distributed energy model like ReOpt.
  • Can a PCM model future system electricity prices?
    • Yes, a PCM can be used to calculate the wholesale Locational Marginal Prices (LMPs) at each node in the power system, based on given fuel prices and demand, solar, wind, and other time-varying forecasts. It is only a forecast or estimate; to understand the uncertainty in the price forecasts, please consider modeling some scenarios that capture the range of uncertainty in the inputs.