Talk:Open energy system models

	This article was reviewed by member(s) of WikiProject Articles for creation. The project works to allow users to contribute quality articles and media files to the encyclopedia and track their progress as they are developed. To participate, please visit the project page for more information.Articles for creationWikipedia:WikiProject Articles for creationTemplate:WikiProject Articles for creationAfC
	This article was accepted from this draft on 22 December 2016 by reviewer Devopam (talk · contribs).

Energy

	Energy portal This article is within the scope of WikiProject Energy, a collaborative effort to improve the coverage of Energy on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.EnergyWikipedia:WikiProject EnergyTemplate:WikiProject Energyenergy
???	This article has not yet received a rating on the project's importance scale.

Computing: Software / Free and open-source software

	This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing
???	This article has not yet received a rating on the project's importance scale.
	This article is supported by WikiProject Software.
	This article is supported by Free and open-source software.

Tip: Anchors are case-sensitive in most browsers.

This article contains broken links to one or more target anchors:

[[power-to-X#sector coupling concepts|sector coupling]]
[[Power-to-X#Sector coupling|coupling]] The anchor (Sector coupling) has been deleted.

The anchors may have been removed, renamed, or are no longer valid. Please fix them by following the link above, checking the page history of the target pages, or updating the links.

Remove this template after the problem is fixed | Report an error

open_eGo: open electricity Grid optimization

This project can be added when it goes live at the end of August 2016. RobbieIanMorrison (talk) 16:39, 21 July 2016 (UTC)

Done. RobbieIanMorrison (talk) 12:05, 26 September 2016 (UTC)

The project seems to have been removed. It was clearly present at one point as indicated. In any case, the project is now known as oGoⁿ. RobbieIanMorrison (talk) 09:11, 18 October 2024 (UTC)

Split page in due course: Open energy systems data

Given that open energy system data projects are on the increase, this page could be split in due course to yield a new Open energy systems data page. RobbieIanMorrison (talk) 12:05, 26 September 2016 (UTC)

Done. Please see Open energy system databases. RobbieIanMorrison (talk) 10:47, 4 February 2017 (UTC)

Spanish wholesale power market data

The following Spanish initiative could be a possible addition to the open data section in due course. Best wishes. RobbieIanMorrison (talk) 20:32, 5 October 2016 (UTC)

"PriceProfor Engine: Iniciativa abierta para el modelado, el análisis y la explotación de datos de precios del Mercado mayorista eléctrico Español" [PriceProfor Engine: open initiative for modeling, analysis, and exploitation of price data from the Spanish wholesale power market] (in Spanish). Retrieved 2016-10-05.

Not done. Shifted comment to the open energy system databases article. RobbieIanMorrison (talk) 10:56, 4 February 2017 (UTC)

Further models

Open energy system modeling projects that could be added in due course
Project	Model type	Academic publications	License details	Code published	Comments
AMIRIS	java-based agent-based modeling framework		Apache 2.0	GitLab	to add
COMANDO	python modeling framework beyond MILP		MIT	GitLab instance	to add
CREST	spreadsheet-based micro-energy system model		CC BY-ND 3.0		cannot add, the ND variant is not an open license
DDPP	spreadsheet-based energy system model	no	not specified
deeco	high-resolution energy system modeling framework		GPLv2		to add (although project abandoned 2005)
Einstein	single-site analysis		GPLv3		to add
EnergyNumbers– Balancing	interactive electricity system model	no	GLPv3
energyRt	model from United States written in R	no	AGPLv3.0		need to confirm license status
ETHOS.FINE	Framework for Integrated Energy System Assessment		MIT	GitHub	to add
FlexTool	Power system flexibility for the energy transition		LGPLv3+	request access	to add
Friendly Sam	energy system optimization toolbox	not known	LGLPv3	GitHub	license needs to be added to codebase
GCAM	Global Change Analysis Model IAM		ECL-2.0	GCAM Core	to add
GENeSYS-MOD	global energy system model (based on OSeMOSYS)		Apache 2.0	GitLab instance	to add
GnuAE	GNU alternative household energy systems	no	GLPv2		to add
GridCal	Research oriented power system software		LGPLv3	GitHub
GridPath	Versatile power-system planning platform		Apache 2.0	GitHub	to add
Grid Sim	browser based public education model		Apache 2.0	GitHub	to add (an initiative by google)
highRES	GAMS-based model for Great Britain and Europe		MIT	GitHub	to add
M-LED	multi-sectoral latent electricity demand platform		MIT	GitHub	to add
mat-dp	material demand projections model		MIT	GitHub	to add
MultiMod	large-scale equilibrium model of global energy markets		not specified	not yet	first top-down model?
OMEGAlpes	district scale model		Apache 2.0	GitLab
open_plan	decentralized systems with emphasis on built environment	no	AGPLv3	GitHub
OpPoDyn	research on the dynamics of networks and power grids		GPLv3	GitHub	to add
PLEXOS Open EU	based on proprietary PLEXOS software		not specified	not yet
PowSyBl	Power System Blocks	no	MPL-2.0	GitHub	to add
Rogeaulito	projected accounting model for global energy		not specified		publication indicates open license, have emailed the developers
Spine	energy system model				to add
SpineOpt	energy system model				to add
StELMOD	stochastic optimization model for European electricity markets		MIT		to add
TIMES Model Generator	framework maintained by ETSAP		GPL‑3.0‑or‑later		to add
WWS project	based at Standford University		MIT		to add
Academic publications: pending or published academic literature or the presence of secondary sources is used as a criteria for inclusion in the article. License details: projects must specify their license first (spreadsheet projects are prone to not doing so). Code published: lists projects that have indicated their intention to make their code public but have not yet done so. Blank cells indicate an affirmative.

Likely AI generated material

Hi - I added the AI generated tag to this article, as the edits by RobbieIanMorrison display several indications of LLM use and thus need extra review for accuracy/hallucinations, sourcing issues, editorializing/non-neutral tone and synthesis, and the like. Gnomingstuff (talk) 22:09, 6 October 2025 (UTC)

@Gnomingstuff: could you expand on that? @RobbieIanMorrison: did you use LLMs to help write some of the stuff in this article?

Let's see if we can work towards removing the tag at some point. —Femke 🐦 (talk) 17:32, 28 November 2025 (UTC)

Absolutely not. I think that tag is not warranted is this case. I have never used Google search to help write text. I have never used ChatGPT at all, ever. I might occasionally run a sentence or two through DeepL Write more recently to see if the text could be better framed. Perhaps someone could give an example of the "offending" material? That would really help me to comment specifically. I should add that I have worked in this field for 30 years. And I have read all the sources I have cited. Interesting times, I guess! Best, RobbieIanMorrison (talk) 18:05, 28 November 2025 (UTC)

And can we compare timestamps? I would guess that most of that article was written prior to LLMs being available. RobbieIanMorrison (talk) 18:07, 28 November 2025 (UTC)

DeepL is a large language model tool. Therefore I have restored the tag, as it is warranted on the grounds of AI use literally being confirmed. "Interesting times" indeed.

The primary visible problem is AI editorializing -- asking AI to "better frame" something is turning over the article prose to synthesis and original research, resulting in such AI-characteristic text like highlighting the need to reduce coal consumption, highlighted the need to coordinate infrastructure policies and import strategies. Which carries several issues:

It's the exact same kind of superficial editorializing that AI produces for virtually any topic (see WP:AISIGNS, "Superficial analyses" and "AI vocabulary" sections)
It is opinion stated in wikivoice; your idea of a "need" may not be my idea of a "need," which is probably not the person reading this's idea of a "need" either. The people holding the opinion, in this case, are the researchers of the study cited. However...
AI-generated text like this tends either to be not backed up by the source at all, or an overly close paraphrase of the source material. The second statement is the latter, as it is almost verbatim from the study cited. And that's just one clause; to "work toward removing the tag," any text written, "better framed," or otherwise touched by LLM tools needs to be similarly scrutinized.

And no, this kind of text was not present in versions prior to widespread AI availability. Gnomingstuff (talk) 14:02, 3 December 2025 (UTC)

Fellow Wikipedia editors: can we gain some perspective please? Firstly:

LLMs can be used in a number of ways, ranging from improving grammar to generating entire texts from a single prompt
my use of DeepL.Write, if indeed deployed when contributing to this article, would have been limited solely as an aid to single sentence formulation

Can someone point to Wikipedia policy that relates to this use‑case of sentence-by-sentence formulation assistance under fine‑grain human agency?

More specifically, relative to this article:

by my estimate, I could have utilized DeepL.Write for the GENeSYS and RAMP sections (but I have no record, of course, of having done that)
by my estimate, I could have utilized DeepL.Write for five or more one sentence edits (the exact number is hard to gauge)
any such usage by me would have been restricted to single sentences under careful control
the change log shows that 84% percent of article commits predate November 2022, a date used by German academics to mark the public uptake of LLMs
I am responding directly because I have authored circa 90% of this article

I want to repeat that I have never used AI to generate text, as normally understood, when editing Wikipedia. Never.

The example in the previous reply regarding the term "highlighting" is bizarre in my view — with each usage backed up by a separate peer‑reviewed article. Those sentences are also separated in the text and do not run together as implied earlier. Both cited publications use the term "highlight*" multiple times (counts: Burandt et al (1 December 2019): 6, Hainsch et al (15 January 2022): 11). The example quoted is indeed the kind of framing that predated LLMs. And, yes, it is acceptable and probably desirable to use phrasings from the original sources when editing articles (without violating copyright of course).

I find the the use of assertions presented as certainties and the reverse burden‑of‑proof evident in this discussion somewhat problematic.

Finally, I did not confirm my use of DeepL.Write when editing this article. That is a misrepresentation that should be corrected (@Gnomingstuff: please note). @Femke: FYI. Best, RobbieIanMorrison (talk) 16:56, 3 December 2025 (UTC)

by my estimate, I could have utilized DeepL.Write for five or more one sentence edits (the exact number is hard to gauge)

I want to repeat that I have never used AI to generate text, as normally understood, when editing Wikipedia. Never.

So which is it, then? Gnomingstuff (talk) 17:05, 3 December 2025 (UTC)

There's a difference between generating completely new text (which has a high hallucination rate), and copyediting (which has a low hallucination rate, and hallucinations are easier to spot as you know what the source says). I do the second sometimes, even though I have no patience for the former. —Femke 🐦 (talk) 17:10, 3 December 2025 (UTC)

There's a difference, but the above comment consists of hundreds of words Gish galloping through minor nitpicks (no shit, those two quotations that don't syntactically fit together are in fact two separate quotations?) while carefully talking around what was and was not done.

As far as the specific issue with the stuff I mentioned (which are only two examples out of who knows how many): If someone states an opinion, that opinion must be attributed to the source, not stated in wikivoice. This is policy, per WP:NPOV: Usually, articles will contain information about the significant opinions that have been expressed about their subjects. However, these opinions should not be stated in Wikipedia's voice. Rather, they should be attributed in the text to particular sources, or where justified, described as widespread views, etc. For example, an article should not state that genocide is an evil action but may state that genocide has been described by John So-and-so as the epitome of human evil. It just so happens that AI violates WP:NPOV in this specific way all the time. Gnomingstuff (talk) 17:16, 3 December 2025 (UTC)

Some finesse is necessary. "As normally understood" refers to generative AI — where a single prompt can produce an entire research report. DeepL.Write is a service that improves grammar and framing under human agency. Very different use cases. I could not have included (quoting from the AI template) "hallucinated information, copyright violations, claims not verified in cited sources, original research, or fictitious references". That is simply not possible under my potential usage and I stress "potential" usage. Therefore the AI tag should be removed because it cannot apply. RobbieIanMorrison (talk) 17:22, 3 December 2025 (UTC)

As someone who does copyedit with AIs, hallucinations do happen in this use case still, even though they tend to be more subtle. AIs occasionally change the meaning of sentences when they copyedit. —Femke 🐦 (talk) 17:24, 3 December 2025 (UTC)

Sure, DeepL.Write can change the meaning. But I check what I use of course. And that is exactly what I am claiming (given that I did use DeepL.Write for the cited example). Best RobbieIanMorrison (talk) 17:32, 3 December 2025 (UTC)

Policy Wikipedia:LLMCOMM "does not apply to using LLMs to refine the expression of one's authentic ideas". That is exactly the case here in the context of possibly a very few edits or perhaps no edits. I am going to remove the AI tag in 24 hours time, unless I hear further arguments to the contrary. Best RobbieIanMorrison (talk) 18:25, 3 December 2025 (UTC)

Three days or so have passed and I am just about to remove that AI tag. Stepping back, the process I experienced as a longtime Wikipedia editor, as documented above, has left a bad taste in my mouth. Wikipedia needs to improve these processes for identifying and rectifying the inappropriate use of generative AI, which I don't doubt is a problem. The apparent current "shoot first and ask questions later" AI policy seems at odds with the normal ethos of Wikipedia. In any case, I found the preceding discussion quite difficult. Best, RobbieIanMorrison (talk) 11:15, 7 December 2025 (UTC)

Done The certain=yes qualifier on the AI-generated template was especially unwarranted.

Again, as a longstanding editor, I would really like to be believed when I make considered statements. Best, RobbieIanMorrison (talk) 11:23, 7 December 2025 (UTC)