Ontology: Is in every energy transformation a material entity participating?

[l-emele]

Description of the issue

Discussion in issue #673 went off-topic as a longish discussion started on whether in every energy transformation a material entity participating. There are different views on whether such an axiom energy transformation has participant some material entity.

Pro:

• The class energy transformation itself is very abstract and hence the axiom must be abstract. But the subclasses are less abstract and hence the derived axioms for the subclasses will be less abstract. The inclusion of the axioms ensures that for every energy transformation one has to think not only about the energy side but also on the material side.

Con:

• The proposed axiom is very abstract. Hence it seems not to be directly useful in the energy modelling domain / in energy economics.

Ideas of solution

If you already have ideas for the solution describe them here

Workflow checklist

• [ ] I discussed the issue with someone else than me before working on a solution
• [ ] I already read the latest version of the workflow for this repository
• [ ] The goal of this ontology is clear to me

I am aware that

• [ ] every entry in the ontology should have a definition
• [ ] classes should arise from concepts rather than from words

[l-emele]

I am not only interested in what opinions of @OpenEnergyPlatform/oeo-domain-expert-energy-modelling are but also
what @OpenEnergyPlatform/oeo-general-expert-formal-ontology think on this topic.

[0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q]

To give the view of one @OpenEnergyPlatform/oeo-domain-expert-energy-modelling:
The material component is inconsequential to energy systems modelling. While it is acknowledged that material entities can carry energy (aka "energy carriers") and thus participate in energy transformations, as in oil → Diesel fuel, ESM does not track masses, only their energy content. So 1 EJ of oil will be converted to 0.91 EJ Diesel fuel, and nobody cares that this means that 23.9 Mt oil go into the refinery and 21.2 Mt Diesel fuel come out. Since ESMs don't track it, is has no business being in the ontology.

[l-emele]

nobody cares that this means that 23.9 Mt oil go into the refinery and 21.2 Mt Diesel fuel come out.

Be careful with generalisations. The statements might be true for your models, but that does not necessarily mean that it is true for all ESM. The model I am working with not only contains energy flows but also contains some material flows associated with energy transformations.

[0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q]

Be careful with generalisations. The statements might be true for your models, but that does not necessarily mean that it is true for all ESM.

To give the view of one @OpenEnergyPlatform/oeo-domain-expert-energy-modelling:

Should I note that I am voicing an opinion for every sentence individually?

---

The model I am working with not only contains energy flows but also contains some material flows associated with energy transformations.

That is – in my opinion – not the same as mass flows being part of energy systems modelling. Our model represents, i.a., population, the labour force, mass flows of 40 GHGs, cement production, water withdrawal and the sort of car people drive. All of which are not part of the ESM. (Explicitly in our model, which is modularised. But, to my mind at least, conceptionally in all models, too.)

nobody cares that this means that 23.9 Mt oil go into the refinery and 21.2 Mt Diesel fuel come out.

The model I am working with not only contains energy flows but also contains some material flows associated with energy transformations.

And these flows are modelled _explicitly_? You have an energy transformation that has some energy carrier as an input or output _both_ in units of energy and units of mass? Because specifying 23.9 Mt oil is equivalent to specifying 1 EJ of oil, assuming (default) specific calorific value.

---

The inclusion of the axioms ensures that for every energy transformation one has to think not only about the energy side but also on the material side.

Is that intended as a normative of merely a positive statement? Because unless you want everybody to engage in mental contortions to figure out some material entity that participates in the energy transformation of an electric motor, _apart from_ the motor, I fail to see the point of this.

The model I am working with not only contains energy flows but also contains some material flows associated with energy transformations.

I'm under the impression that according to the suggested axiom, you have some thinking to do to come up with material flows for the _other_ energy transformations in your model.

---

Poking around in the ontology, I can't help but notice that all kinds of energy converting devices participate in some energy transformation. I would expect participates in to be the inverse of has participant – which it is not. has participant has inputs and outputs as subclasses, participates in has input and output of (something). So it seems to me that these relations don't work for energy converting devices in either direction and need extension or additional relations are needed.

There are different views on whether such an axiom energy transformation has participant some material entity. [_sic_]

Conditional on how the "input/output" v "device the magic happens in" and has participant inverse of participates in issues shake out, this axiom would already be fulfilled by many (didn't check all) energy converting devices, which are also material entities.
In my opinion.

[0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q]

While I consider the reproach of undue generalisation unwarranted, since I spoke specifically about my opinion, I do acknowledge that my response in turn was overly harsh, disparaging, and uncalled-for, and thus in violation of the project's code of conduct. I hope that @l-emele accepts my sincere apology and that others who might have been discouraged from participating in this discussion do so too.

I encourage @OpenEnergyPlatform/oeo-domain-expert-energy-modelling and @OpenEnergyPlatform/oeo-general-expert-formal-ontology to take up this thread again and contribute their opinions. I will refrain from commenting until it goes into a stage of discussion, at which point I will strive to put my views forward in an appropriate manner.

[l-emele]

@0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q: I do accept your apology and I am looking forward to constructive discussions in future.

[sfluegel05]

So the state of this issue as I see it is currently: Some models use material entities, some models don't. Everyone agrees that there always some material entities participating in an energy transformation (which would be the answer to the question from the issue title). However, not everyone agrees on whether it is a good idea to include it as an axiom in the ontology.
Pro: you have to think about the material entities participating in the transformation (-> better axiomatization)
Con: you have to think about the material entities participating in the transformation (-> more effort)

Maybe it helps to clear up what kind of material entities we are talking about. This issued originated from #673, where the use case was that syngas (or similar products) are the output of a solar chemical energy transformation. But then the discussion shifted from has output to has participant (see https://github.com/OpenEnergyPlatform/ontology/issues/673#issuecomment-795620183).
We should distinguish between

1. the energy carrier(s) which are input / output of an energy transformation, i.e are modified or produced (e.g. syngas)
2. the energy carrier(s) which carry the energy which is input / output of an energy transformation, but remain unchanged themselves (e.g. batteries)
3. the participants which are needed to perform the transformation but don't carry any energy (e.g. PV cells)

The discussion has shown that (1) doesn't always exists. Therefore the proposed axiom is only talking about has participant and not has input / output. I would guess that (2) doesn't always exist either (e.g. for solar chemical energy transformation, the input-photons don't exist after the process and the output-syngas didn't exists before the process). But one of (1) and (2) always exists, i.e. there is always an energy carrier. I'm not sure about (3), but it seems to me that this group is not the type of material entities intended by the issue.

My conclusion: We should include the axiom energy transformation has participant some material entity, but that doesn't mean modellers have to explicitly state material flows for every energy transformation (which would be category (1)). E.g. for an electric motor it would suffice to say that there is a battery (category (2)) that is participating in the process. Making such relations seems rather trivial to me for the energy transformation-subclasses we have.