Renewable Energy Transition Modelling

Modelling
, , ,
1

Hello,

As of late I got more interested in modelling energy systems (see below for some links). In particular, I am interested in numbers for a global Renewable Energy Transition. Therefore, may I ask if there is a simple but global model for said transition available somewhere in order to get an idea of the orders of magnitudes of some quantities involved, a model that does nevertheless take into account some essential aspects:

  • phasing out fossil fuels according to a target function (open loop, no policy),
  • efficiency gain (and hence decreasing total energy need) in industrial processes, domestic settings or mobility through electrification,
  • non-renewable resource requirements (copper etc) with decreasing ore quality and energy investment into renewable infrastructure, both for the transition phase and a later stage with moderate growth,
  • hydrogen need (from grey to green) to maintain ammonia consumption at least at same level

The question I would like to answer is: what will be the collateral damage of the renewable energy transition.

Kind regards,
Peter Wurmsdobler
Cambridge, UK

https://peter-wurmsdobler.medium.com/
https://www.linkedin.com/in/peter-wurmsdobler-b87663/

Some energy related posts:

2

Hello @PeterWurmsdobler

Some projects have developed educational versions of their frameworks in the sense that the codebase remains but the models themselves are intentionally simplified. You may find some pointers here:

But be aware that training models sometimes deploy fictitious countries, so not really what you want.

This reduced but not educational use of the PyPSA framework could be worth watching for some hints:

There are some global models too. OSeMOSYS has been applied in this context. The work of UK think tank TransitionZero could be worth reviewing in relation to your inquiries â€” including for global data.

Some researchers are starting to move into the industrial sector because they need to cover industry feedstocks, including carbon dioxide: https://forum.openmod.org/tag/industrial-ecology The PyPSA project would be one such community, also looking at methanol in particular.

Lots of projects include green hydrogen these days.

There is also a partial list of open models on Wikipedia that you might find of help too when hunting around.

I am sure that any of these projects would be happy to have you on board to extend or develop these more simplistic models. Of course, most early stage researchers need to make things as complex as possible â€” and they do make up the bulk of the modeling workforce.

Perhaps your research question should be: what is the collateral damage of not doing the net‑zero transition?

hope this helps … and welcome, R

PS: Did you see AntĂłnio Guterres quoted in The Guardian today:

Guterres added: “To tackle all these symptoms, we need to fight the disease. The disease is the madness of incinerating our only home. The disease is the addiction to fossil fuels. The disease is climate inaction.”

â–˘

1 Like
3

Hello @robbie.morrison,

many thanks for the leads to various projects; I will try to study them in more detail.

Being an engineer, I usually like to start with simple models first, as simple as they can be, lumped mass models, to get an idea of the orders of magnitude involved, then refine as needed. I made a mistake early in my career modelling systems to such a detail and complexity that the outcome was not easy to interpret any more and contained many modelling artefacts.

I think that not going for a net-zero transition is not even an option to be considered. Only I have an inkling that the net-zero transition is only a tad better for planet Earth.

What Guterres says is not wrong, but not sufficient either, IMHO. CO2 emissions from burning fossil fuels resulting in climate change is the tip of an iceberg; the underlying growth based economy is still not being questioned.

Cheers, peter.

1 Like
4

Post scriptum, to give an idea of my general way of thinking, I like to stake out the ranges of quantities first before doing any detailed modelling.

For instance, suppose the current energy demand is about 20TW on average, mostly from fossil fuels. Electrification of transport, heating etc will result in needing less, perhaps half (thermodynamic efficiency of combustion engine being 25% - 30%, combined cycle gas turbines perhaps even 60%). So about 10TW average electric power would be needed. This is good news.

Imagine if all that energy came from solar power post-transition, with panels distributed all over the world and UHVDC lines creating a global well balanced grid. I would hope to get about 20W/m2 out of every for a solar panel, i.e. 10 * 10^12 / 20 = 0.5 * 10^12 m^2 installed panels needed, or about 0.5 million km^2. Just for the order of magnitude. If half of the 10TW comes from solar panels, well, less is needed, if the grid is less balanced, then more. But the order of magnitude is about 1 million km^2.

Now, what about materials. I haven’t done the research on that, but I could well imagine to allow 200g/m2 silicon, 1kg for glass, 1kg for aluminium or steel structure, not even talking about inverters, grid and other infrastructure. If I know the extent of materials needed for the transition, I can look into more detail what energy is needed to produce that in the first place during the transition, i.e. additional power capacity. In a post-transition world, assuming a life time of say 20 years per panel, and a recycling rate of say 90%, how much virgin material (million tons of copper, etc) will still be required per year, and is that even possible? This is the kind of question I would like to be able to answer.

This process should yield some overall figures to work out the feasibility of the transition as well as a world post-transition. It would even be more fun to add some dynamics to that, with some differential equation and feedback, and then extend the model.

Cheers, peter.

5

Some sources on raw materials. The IEA report is good:

And this book was written by researchers in Germany:

  • Wellmer, Friedrich-W, Peter Buchholz, Jens Gutzmer, Christian HagelĂĽken, Peter Herzig, Ralf Littke, and Rudolf K Thauer (2019). Raw materials for future energy supply. Cham, Switzerland: Springer. ISBN 978-3-319-91228-8. Hardback. Translation of German Rohstoffe report.

For a jump start, try the Material Demand Projections Model (or Mat‑dp model), licensed MIT, written in Python, and developed at the University of Cambridge:

It looks like the data is embedded in the code?

Good luck with your systems dynamics modeling too!

â–˘

1 Like
6

Just adding that the TransitionZero work Robbie mentions builds on OSeMOSYS Global: GitHub - OSeMOSYS/osemosys_global: A global power system model generator for OSeMOSYS. Documentation at Welcome to the documentation of OSeMOSYS Global!. OG is a global electricity model generator.

Cheers,
Taco…

2 Likes
7

Adding a recent preprint from the PyPSA‑Earth project (this is the model generator, the wider initiative is known as PyPSA meets Earth):

1 Like
8

You might be interested in GLUCOSE. It is a simple global climate-land-energy water model which is built using OSeMOSYS. You can read the paper here and download the model from Github.

The GLUCOSE model is: “a highly-aggregated global IAM, open and accessible from source to solver and using the OSeMOSYS tool and the CLEWs framework. The model enables the exploration of policy measures on the future development of the integrated resource system. Thanks to its relatively simple structure, it requires low computational resources allowing for the generation of a large number of scenarios or to quickly conduct preliminary investigations. GLUCOSE is targeted towards education and training purposes by a range of interested parties, from students to stakeholders and decision-makers, to explore possible future pathways towards the sustainable management of global resources.”

1 Like
9

That citation in full:

1 Like
10

This dashboard for Paris Compatible (PAC) scenarios for Europe in 2030 may be helpful. Results at the country level can also be displayed:

  • RGI and RLI (2023). CLEVER (2030) Dash. Renewables Grid Initiative and Reiner Lemoine Institute. Berlin, Germany. Interactive website.

Perhaps there are similar initiatives for the whole planet? A number of teams have also modeled European PAC scenarios and a simple web search should turn most of these up.

1 Like