Energy carrier in addition to energy carrier disposition

l-emele commented 4 years ago

Description of the issue

In the ontology paper today we discussed that we might have distanced the OEO from the energy modelling experts with having only an energy carrier disposition but not a class energy carrier. Energy modelling expert would usually look for energy carrier and not for the disposition.

The current def of energy carrier disposition in: An energy carrier disposition is a disposition of an object or object aggregate that contains energy for conversion as usable energy.

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 commented 4 years ago

@fabianneuhaus and @tillmo : You had during the telco already some ideas for solution but I can't remember the details. Could you please write down your idea?

tillmo commented 4 years ago

This would be a class with formal definition (i.e. EquivalentTo) has_disposition some energy_carrier_disposition.

stap-m commented 4 years ago

This would be a class with formal definition (i.e. EquivalentTo) has_disposition some energy_carrier_disposition.

@tillmo @fabianneuhaus what would be the parent class? The energy carrier disposition, as is defined, is applicable to portions of matter (fuels) and artificial objects (e.g. storage units).

stap-m commented 4 years ago

Btw: the subclasses of energy carrier disposition are not defined as dispositions but as actual "energy carriers". (I guess we didn't adjust the defs properly). We should adjust them in the following!

tillmo commented 4 years ago

If you look at classes which can be inferred to be subclasses of this newly defined class energy carrier, you get air, coal, peat, water, biogas, biofuel, methane and uranium. I think the parent class should therefore be portion of matter.

l-emele commented 4 years ago

So matter and energy carrier are equivalent and we ontologically rediscovered Einstein's mass-energy equivalence. :smile:

stap-m commented 4 years ago

If you look at classes which can be inferred to be subclasses of this newly defined class energy carrier, you get air, coal, peat, water, biogas, biofuel, methane and uranium. I think the parent class should therefore be portion of matter.

Yes, but the current state is not complete in my opinion.

Taking a look at the subclasses of energy carrier disposition, there is energy storage aswell. Here, the hierarchy is obviously wrong: battery storage is not an energy carrier disposition but can posess it, therefore shoud get the relation has_disposition some energy_carrier_disposition and needs to be reclassified as artificial object / energy storage object.

Maybe we coud split it up into several "energy carrier classes": e.g. for the portions of matter that have also the fuel_role we already have the equiv class fuel.

tillmo commented 4 years ago

With the parent class included, the definition is

energy_carrier EquivalentTo: portion_of_matter and has_disposition some energy_carrier_disposition

This then implies that fuel is a subclass of energy_carrier (i.e. as part of the inferred subclass hierarchy).

l-emele commented 4 years ago

What would be a good definition for energy carrier? First suggestion, similar to the disposition definition: An energy carrier is a portion of matter energy that contains energy for conversion as usable energy.

stap-m commented 4 years ago

Sorry, I still disagree, that energy carrier just refers to portions of matter. If such a class is favoured we have to either

specify it as e.g. energy carrying portion of matter (very ambiguous though)
OR kick out the energy storage-part of the energy carrier disposition

l-emele commented 4 years ago

To be honest, the difference between energy storage and energy carrier disposition is not really clear to me:

Energy storage is a function of an artificial object that has been engineered to contain energy for conversion as usable energy later.
An energy carrier disposition is a disposition of an object or object aggregate that contains energy for conversion as usable energy.

So basically the difference is only that one is a disposition and the other is a disposition?

stap-m commented 4 years ago

and energy storages refer only to artificial objects.

It's there twice btw, as function and as disposition. Oops! grafik

edit: apparently, it is not a bug, but implemented on purpose, see this comment by @jannahastings in #209

BFO 'function' is a subclass of 'disposition' because functions are 'good', selected dispositions. Thus, it makes sense to me that 'EnergyStorage' can be a subclass of 'EnergyCarrier' and at the same time be a function. In practice, that will have to be asserted as two parents, but nevertheless it will not create any inconsistencies and I think it is correct in this case.

How about creating the additional class EnergyStorageObject beneath ArtificialObject, defined as 'an artificial object that has the function EnergyStorage' and classify battery, etc. beneath that?

tillmo commented 4 years ago

I agree that the difference between energy storage and energy carrier disposition is not really clear. We should add Wikipedia's definition "Energy storage is the capture of energy produced at one time for use at a later time."

Concerning energy storage occurring twice, not that this is still one class, just one with two parents. I think this is not a problem. However, we could avoid this by letting energy carrier disposition be a subclass of function, too (if this is appropriate).

Concerning having energy carrier as a subclass of portion of matter, Wikipedia says, "According to ISO 13600, an energy carrier is either a substance or a phenomenon that can be used to produce mechanical work or heat or to operate chemical or physical processes." So maybe engery carrier should a subclass of portion of matter union something else?

tillmo commented 4 years ago

I think engery carrier should a subclass of portion of matter union artificial object. Then battery etc. is included.

l-emele commented 4 years ago

Maybe it is enough to simply use "energy carrier" as synonym for "energy carrier disposition"?

akleinau commented 4 years ago

Maybe it is enough to simply use "energy carrier" as synonym for "energy carrier disposition"?

Is this disposition what energy modelling experts mean when they say "energy carrier"? So when an energy modelling expert says "energy carrier", does he or she mean both portions of matter and batteries?

l-emele commented 4 years ago

For me a battery is not an energy carrier but an energy storage object. The chemicals in the battery are the energy carrier in that case. Similar with a pumped hydro storage: The plant itself is the storage object but the water contained in the reservoir (which is part of the pumped hydro storage) is the energy carrier.

akleinau commented 4 years ago

than just making it a synonym would include things like battery and therefore not be the wanted meaning. When I understand correctly @stap-m disagrees with that, can you ask more people to find out whats the more common understanding? Just portion of matter or also batteries?

l-emele commented 4 years ago

I think we should discuss this on our next meeting.

stap-m commented 4 years ago

For me a battery is not an energy carrier but an energy storage object. The chemicals in the battery are the energy carrier in that case. Similar with a pumped hydro storage: The plant itself is the storage object but the water contained in the reservoir (which is part of the pumped hydro storage) is the energy carrier.

I'd actually agree with you: the battery contains the energy carrier but does not neccessarily be it. Our def of energy carrier disposition isn't explicit here, though. If we changed the def such that it just refers to portion of matter, this might solve the problem. Then, we'd also have to reclassify energy storage, which is a subclass of energy carrier disposition currently.

0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q commented 4 years ago

So when an energy modelling expert says "energy carrier", does he or she mean both portions of matter and batteries?

For me, energy carriers are used to carry energy some place else. Coal, oil, gas. Also methanol, hydrogen, and other funky stuff. Batteries don't contain energy carriers, as batteries are not used to carry energy some place else (with some exceptions), and certainly battery acid is not used to carry energy anywhere, but to have energy wherever one happens to be. "Air" and "water" as energy carriers are really stretching it. Yes, they contain energy (in some sense). But where is geothermal energy contained? So add "rock". Ambient heat pumps … add everything around you, as it will get (very slightly) cooler when using one.

So, for my work, I care about energy carriers when they are traded. Else, they are a short-hand description for energy transformation processes that mainly use certain energy carriers, i.e. "primary energy uranium" would be the amount of "nuclear energy" extracted (from nature) and turned into electricity.

l-emele commented 4 years ago

Energy carrier disposition is currently defined as: An energy carrier disposition is a disposition of an object or object aggregate that contains energy for conversion as usable energy.

The concept you are describing is in the OEO more a fuel which is defined as: A fuel is a portion of matter that has the disposition to be an energy carrier and which has a fuel role that is realised in processes that release the carried energy by transforming the portion of matter into a different kind of portion of matter in a way that releases heat or does work.

Vera-IER commented 4 years ago

So, for my work, I care about energy carriers when they are traded.

I agree to that. In my understanding, energy carriers are fuels (like for example gasoline) and also primary energy carriers (like crude oil or biomass). I would not consider batteries or storages as energy carriers.

This definition fits to my understanding: (source) p.3

Energy Carriers. Energy carriers (sometimes called energy currencies) are the energy forms that we transport and use, and include some energy resources (e.g., fossil fuels) and processed (or secondary) energy forms (e.g., gasoline, electricity, work and heat). The processed energy forms are not found in the environment.

(However I would also not consider electricity, work and heat as energy carriers.)

With the def of energy carriers from the withdrawn ISO, batteries could also be energy carriers:

Substance or phenomenon that can be used to produce mechanical work or heat or to operate chemical or physical processes.

0UmfHxcvx5J7JoaOhFSs5mncnisTJJ6q commented 4 years ago

However I would also not consider electricity, work and heat as energy carriers.

I very much do consider electricity and heat to be energy carriers (within the systematics of our model), as they are traded.

With the def of energy carriers from the withdrawn ISO, batteries could also be energy carriers:

Substance or phenomenon that can be used to produce mechanical work or heat or to operate chemical or physical processes.

I can see why this was withdrawn. I would challenge you to find anything that is not an energy carrier under this definition. ;)

Vera-IER commented 4 years ago

I very much do consider electricity and heat to be energy carriers (within the systematics of our model), as they are traded.

Yes I agree that in this sense, electricity and heat could also be seen as energy carriers.

I did found another def/explanation of energy carrier from IPCC:

Energy carriers include electricity and heat as well as solid, liquid and gaseous fuels. They occupy intermediate steps in the energy-supply chain between primary sources and end-use applications. An energy carrier is thus a transmitter of energy.

And they describe it with that picture: energy carrier

l-emele commented 4 years ago

As discussed bilateral with @stap-m we move all energy (carrier) related issues to v.1.3 as this needs more discussion in a dev meeting.

sfluegel05 commented 3 years ago

It seems to me like we have to answer two questions here:

1. What are energy carriers / Who should have an energy carrier disposition?

The answer so far is:

artificial objects like batteries are not energy carriers
portions of matter can be energy carriers
~energy~electricity and heat can also be energy carriers

2. Do we need an energy carrier class?

The answer so far:

for energy modelers, the concept energy carrier is more intuitive than energy carrier disposition
idea 1: add an equivalent class energy carrier defined as EquivalentTo has_disposition some energy_carrier_disposition
idea 2: include the term energy carrier as a synonym of energy carrier disposition

Assuming we agree on portions of matter, energy and heat for question 1, I am in favor of idea 2, because the discussion has shown that it is hard to find a subclass of energy carrier since the concepts which would become subclasses of it (namely portions of matter, energy and heat) are vastly different. So it makes sense to keep using the energy carrier disposition, which describes a characteristic of certain concepts (i.e. their ability to carry energy).

stap-m commented 3 years ago

We have too many posts here (28). I'm preparing a summary for this issue, so that we can discuss it in the next meeting.

stap-m commented 3 years ago

From the dev meeting #12: new agreed definition: An energy carrier disposition is a disposition of an material entity that contains energy for conversion as usable energy.

idea: add anonymous classes:

The case for heat: ('material entity' and bearer_of some 'thermal energy') has_disposition some 'energy carrier disposition' (anonymous class "a hot thing")
The case for electricity: ('material entity' and bearer_of some 'electrical energy') has_disposition some 'energy carrier disposition' (anonymous class "a charged thing")
The case for potential energy: ('material entity' and bearer_of some 'potential energy') has_disposition some 'energy carrier disposition' (anonymous class "an elevated thing")
The case for kinetic energy: ('material entity' and bearer_of some 'kinetic energy') has_disposition some 'energy carrier disposition' (anonymous class "a moving thing") @jannahastings should we actually create those anonymous classes or were they just examples for us and solve it via axioms of energy?

proposal for a def: 'energy carrier' equivalentTo 'material entity' and has_disposition some 'energy carrier disposition'

Or we may link every energy to some carrier: energy inverse(bearer_of) some (material entity and has_disposition some energy carrier disposition)

Subclasses stay for the moment, we should discuss this later...

jannahastings commented 3 years ago

* The case for kinetic energy: ('material entity' and bearer_of some 'kinetic energy') has_disposition some 'energy carrier disposition' (anonymous class "a moving thing")
  @jannahastings should we actually create those anonymous classes or were they just examples for us and solve it via axioms of energy?

We can actually create them using the facility for "General class axioms" at the ontology level. For this we need to specify slightly differently to have the correct syntax. For the above example this would be specified as:

('material entity' and 'bearer of' some 'kinetic energy') subClassOf 'has disposition' some 'energy carrier disposition'

jannahastings commented 3 years ago

sfluegel05 commented 3 years ago

Thanks for the explanation. Assuming we agree on using general class axioms, this should be ready for implementation.

stap-m commented 3 years ago

I am openting a branch and start the implementation.

OpenEnergyPlatform / ontology