PyPSA question: how to model links as H2 pipelines

[Toni Vayá Soler]

Dear Fabian, PyPSA community,

I am a relatively new user of PyPSA and I would like know more about how to model hydrogen value chains with PyPSA to compute optimal production and delivery (i.e. distribution and storage) costs of hydrogen.

Here is simple model I am working on that I would like to refine and expand. It consists of 3 buses and 2 links:

Bus 1: electricity bus with all the electricity generators

Link12: link working as a electrolyser

Bus 2: hydrogen bus. This bus could also include other "hydrogen generators" such as chemical reformers

Link23: link working as a hydrogen pipeline

Bus3: end-user bus, including an store element and an hydrogen load

My question is: is it possible to make Link23  work as an extendable pipeline that PyPSA can optimize. If yes, how?

I am not sure how PyPSA will use inputs such as capital cost and length in this case. If I introduce capital cost in currency/km, PyPSA will simply multiply capital cost by length?

If this topic has already been discussed here do not hesitate to point me to other conversations. The same if there are open PyPSA examples out here that I can consult. Any advice on modelling electrolysers with links is also welcome.

Thank you in advance for your help.

Antonio

[Paul Reggentin]

Hi Antonio,

The PyPSA-Eur-Sec model has a functional H2 network implemented, with various generation and consumption modes as well as an extendable H2 network. If you download the code, the relevant sections are in the file prepare_sector_network.py. It might be a bit more complicated than your application, but you should be able to get some useful information from it.

I haven't looked in detail into it, but the code doesn't seem too complicated, mostly just adding H2-specific buses, links, stores and carriers.

Best,

Paul

[Johannes Hampp]

Dear Antonio,

Yes, you can use a regular link for your "Link23" as pipeline
substitute. It'll be a simple transport model, no pipeline specific
properties implemented.

For the link properties, see:
https://pypsa.readthedocs.io/en/latest/components.html#link

* capital_cost are specified per unit of p_nom
* The length attribute is not automatically read or used to scale the
capital_cost, but just an helpful auxiliary attribute which can be
stored with the link

For pipelines you want to scale capital_cost and efficiency with the
length. Depending on how you consider compressor cost and energy demnand
those either affect capital_cost and efficiency or add additional
marginal_cost.

As Paul hinted you can find the same approach applied in e.g.
PyPSA-EUR-Sec, cf.
https://github.com/PyPSA/pypsa-eur-sec/blob/8dd143f369e8a5f25708500139fa993b0bb173ae/scripts/prepare_sector_network.py#L1025-L1037


Best,
Johannes

Best regards,
Johannes Hampp (he/him)

Justus Liebig University Giessen (JLU)
Center for international Development and Environmental Research (ZEU)

mailto: johanne...@zeu.uni-giessen.de

Senckenbergstr. 3
DE-35392 Giessen
https://uni-giessen.de/zeu

Am 06/08/2022 um 19:59 schrieb Paul Reggentin:
> Hi Antonio,
>
> The PyPSA-Eur-Sec
> <https://pypsa-eur-sec.readthedocs.io/en/latest/installation.html> model

> --
> You received this message because you are subscribed to the Google
> Groups "pypsa" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to pypsa+un...@googlegroups.com
> <mailto:pypsa+un...@googlegroups.com>.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/pypsa/30e629d2-5eed-4328-9faa-7528a596c60dn%40googlegroups.com
> <https://groups.google.com/d/msgid/pypsa/30e629d2-5eed-4328-9faa-7528a596c60dn%40googlegroups.com?utm_medium=email&utm_source=footer>.