Please make suggestions for do-a-thon topics/sessions at the workshop at NREL in September 2019
I intend on attending (as probably the only African representative - South Africa more specifically).
Aside from that being the case, I think there may be interest in exploring the following under modelling framework choices for long-term national/regional level energy planning in developing countries:
- What process is typically followed to choose a modelling framework(s)?
- Is more than one a good idea?
- Who chooses the energy modelling framework(s)? (local institutions, domestic/international consultants, development funding agencies)
- What are the key dimensions to consider when choosing an energy modelling framework(s)?
- What are the typical skill sets needed to support maintenance and updating of established energy models? i.e. ensuring longevity.
- What infrastructure (hardware/software) is needed to run increasingly complex long-term energy models?
I understand that some of these items may have already been answered generally but the application towards developing countries/regions is where I would like the focus to be.
@jarry7 The following document might be of interest then.
- Anon (May 2019). Key principles for improving the support to strategic energy planning in developing and emerging economies. Original filename suggests version 4 and final.
Quoting the last principle:
Transparency and accessibility. Promote open access to and review of planning inputs (data, model design and assumptions) and encourage the accessibility of planning outputs to key stakeholders, subject to government restrictions and commercial confidentiality constraints.
That framing is not especially satisfactory but it is a step in the right direction.
As far as I am aware, the document is not available on the web. Nor do I know who produced it.
Thanks for this.
This was discussed and developed at the EMP-A session at the beginning of this year in Cape Town by a range of stakeholders and contributors (see the end of the document).
It is useful and was developed by a roundtable format dialogue on one of the last days.
Thanks for sharing in any case.
I think there could be value in a do-a-thon focused on the availability of clean, public input data sets used in our models. This is in line with the Open Data forum on this site , but could be beneficial to survey workshop attendees about:
- the resources they are using,
- are resources available publicly,
- if additional work is required to extend the flexibility of a data set to a new application,
- and who is working on what and expected timelines.
Along these lines, some colleagues and I have been working to clean the raw EIA Hourly Demand Data at balancing authority level and push this to an open access databases. From what we see, we are not repeating prior publicly available work.
I would be interested in discussing inter-model comparisons for electricity systems models. Some key questions include:
- What inputs are necessary for all models (fixed and variable costs, high-resolution demand data, etc.)?
- What types of outputs can we compare (e.g., electricity dispatch [at what level?], system costs, system reliability)?
- What parameters should be simplified or ignored in models?
- How do we simplify components for comparison?
The focus of this would be to produce a set of inputs that can be applied to various models for comparison.
This topic can be expanded in a number of ways and I am open to suggestions for other key questions.
Hi @CandiseH In case you are not aware, there is already traffic on this forum on the topic of model comparisons:
There is also at least one model comparison exercise well underway in Europe:
In addition, the following organizations have been talking about running comparison projects too, but I don’t know the current state of their work (others might):
- Berlin‑based Reiner Lemoine Institute (RLI)
- the Energy Modelling Platform for Europe (EMP‑E) network
If you need to translate from German, deepl normally does a good job.
Thanks for posting these! Our interest is more in generating in a do-a-thon – and then make publicly available – a simple and generalized set of inputs that can immediately be used by all (or at least most) electricity dispatch modeling frameworks, regardless of open/closed source, level of complexity/simplicity, etc., to compare with other models. I would be interested to see what the Open Source Energiewende project produces that can be used by all groups.
I’d like to suggest a do-a-thon on creating an open source resource planning modeling platform (i.e., capacity expansion, production cost, etc.).
My team primarily works on regulatory policy and in many of the proceedings where we represent a client (e.g., consumer advocates, PUC, environmental NGO, emerging tech company or industry group), we are often faced with situations such as a utility proposing a new gas plant using justification from a black-box and proprietary resource planning model.
Oftentimes, these models/software do not accurately represent the constraints of new technologies like battery storage and thus cannot employ them in the model’s optimization framework which generally leads to an underestimation of the resource’s value.
In regulatory proceedings, stakeholders are often forced to sign NDAs to view the underlying inputs and outputs of these models. Not to mention, how the model actually works (e.g., sourcecode) can’t even be shared. This goes against intention of having open and transparent processes for incredibly important matters like building 30-year investments that ratepayers fund, especially so if these investments are for fossil fuel resources.
Due to the lack of transparency, stakeholders in utility proceedings often spend countless hours of resource time (internal and external consultants, which cost incremental $) trying to administer and interpret data requests and understand the utility’s analysis and modeling to determine if their story is legitimate.
Because so many models/software that utilities use for planning are black boxes, many stakeholders employ their own consultant to undertake their own modeling exercise. For example, in XCEL’s IRP in MN, there are 4 different models being employed by the various stakeholders intervening in the proceeding. This creates a lot of confusion for decision-makers such as Commissioners with respect to whose analysis they should trust.
As a result, many states with stakeholders who can’t afford expert consultants to help them analyze utility proposals end up with resource investment decisions that are not optimal, especially given the need to decarbonize. This is also true for utilities and governments in developing countries.
Goals and Objectives for a Resource Planning Modeling Platform:
- Platform is open and transparent.
- Platform is free or very affordable so that all policy makers can use.
- Platform adheres to standards so that regulatory agencies across the world can use and trust the quality of the model.
- Platform is continually updated and tested with the latest research and methodological updates regarding resource planning modeling best practices and for new technologies
- Platform has a consistent source of funding so that it can be maintained
I am not proposing we build one from scratch. There are so many great tools out there with software engineers who’d be willing to open-source. I also know that there have been many attempts at undertaking such a project. The real need I see is that in order for such a platform to be adopted in enough regions to have network effects, we would need to define and develop a framework/governance structure for building a community to support the platform. I would love your question, comments, and feedback.
I’ll be prepared to present my team’s research on the different tools out there and how they are being employed in the different states across the U.S.
All the best,
USENSYS (US ENergy SYStem simulation and optimization model - with R and energyRt) is an ongoing initiative to develop a capacity expansion model for US, starting with electric power sector, with the main focus on renewables and transition to zero-carbon economy.
energyRt is an energy-modeling library written in R (R-package) which facilitates development and application of capacity expansion, technological models.
Potential structure of the hands-on session:
- energyRt main features and examples (CHN_ELC_PRO, USENSYS, and more);
- installation (R, RStudio, required packages, GLPK and/or GAMS, energyRt and USENSYS);
- UTOPIA model with energyRt;
- USENSYS balancing version (49 region, 1 full year, 8760 hours version);
- USENSYS transition version (49 region, 2015-2050+ version with flexible sub-annual time-slices);
- analysis and visualization of results;
- further steps, demand for contribution to the model and the package development.
OpenMod Do-A-Thon: Improving Open Energy Data Resources & Documentation
OpenMod have an excellent wiki (https://wiki.openmod-initiative.org/wiki/Data) set up for energy data but there’s a great deal we can still contribute as a community. We are looking for do-a-thon-ers interested in helping build this OpenMod resource and helping others understand the feats and foibles different US energy data sources can offer.
We’d like to kick off the do-a-thon with an intro from the Catalyst Cooperative, RTI International, and Carbon Impact Consulting on work we’re doing to improve the data foundation for electricity modeling. We’d describe what we’ve been able to accomplish with our data and outline a better data wish list. Then the do-a-thon - tell the energy modeling community everything you know about what data are available and what we can do with them. We’d close by assembling a list of open energy data development priorities for the coming year(s).
For each dataset add:
- What are the key variables or dimensions of interest, why is it useful.
- What do you love, hate, question about the data
- How have the data been used
A few options for datasets to get us started:
EIA: Forms 923, 860, 861
EPA: CEMS, eGRID, NEEDS,
FERC: Form 714, Form 1
Based on feedback from NREL workshop registrants, I’ve created a google sheet to track do-a-thons, gauge interest, and enable discussion.
I’ll also keep tracking this thread to keep the sheet up to date.
I’ve also created a survey form to help get feedback and identify participation in the proposed do-a-thons. Please fill out the survey.