Using the Unit Of Measure ontology to distinguish between energy production and consumption.

[bnouwt]

Hi,

in our project we are exchanging measurements/observations about electric energy production and consumption. We are using OWL ontologies and RDF to describe the semantics of these observations and would like to reuse the Unit of Measure ontology for providing the Property that is being measured. We have three questions about this:

1. We cannot find any instances (quantities, or other) in the Unit of Measure ontology that we could reuse to distinguish between electrical energy production and consumption. Is this correct or are there ways to describe this distinction, but we just were not able to locate them?
2. If it is not possible to express this distinction in the current version, would it be possible to include this distinction in a future version or does it not fit within the scope of the ontology?
3. If the distinction will not be in a future version, how do you advise us to model our own instances of production and consumption properties and link them correctly to your ontology?

Hopefully you can help us out with these questions.

Thanks in advance!

Barry

[HajoRijgersberg]

Hi,

Thanx for your interest in OM and your good questions. Indeed, we have not yet defined defined specific electrical quantity subclasses for/on production and consumption. Which exact electrical quantity subclasses would you like to be defined? Then I'll do that. In principle you may decide to do that yourself in a private extension of OM, but if we would choose for the above option (definition by me in OM itself), then other users could use these newly-defined quantities too and that would be better from a standardization point of view. Looking forward to your response. Thanx in advance! :)

Cheers, Hajo

[GeorgJung]

Hi @bnouwt, I have a feeling you might be with TNO? Never mind though. I know exactly what you want to accomplish here, and it is generally a good idea. However, the use of a quantity (produced, consumed, average, minimum, maximum, boundary, etc.) is not part of the unit. This information should be in another ontology. We obtained this insight the hard way in the InterConnect project. Here is a maybe somewhat far-fetched comparison: When a glass contains 100ml, is it half full or half empty? Should that be in the unit?

[bnouwt]

Hi,

@hajo: thanks for your swift response :thumbsup:. @georg: TNO indeed :wave:. Thanks for tuning in!

We fully agree that it is better to reuse concepts instead of defining them ourselves. However, we are unsure which exact electrical quantity subclasses we would like to have defined and can use some guidance in that regard.

As mentioned above, we are mapping energy (kWh) production and consumption measurements (15 minutes resolution) from an API provided by the grid connection provider. We would like the RDF to capture the following things:

• The measured value (i.e. 40)
• The unit of this value (i.e. kWh)
• Whether it is energy or power
  • If we understand correctly, energy is measured in Watt hour and power is maesured in Watt.
• Whether it is consumption or production.

In principle we could encode production as negative (-40) and consumption as positive (+40) measured values, but we rather have this distinction more explicit. So, what would be the best way to include this information? Since I created this issue I had some discussion internally and was pointed to the SAREF 4 Energy property called s4ener:hasUsage with which we would be able to distinguish between production and consumption.

@georg: Was this what you meant when you said it should not be part of the unit and should be in a separate ontology? Also, I think I understand your point with the glass is half-full/half-empty; we need to draw a line somewhere where we stop calling something as being part of the unit.

@hajo: I would be curious about your opinion on this?

Best regards,

Barry

[GeorgJung]

Hi Barry (@bnouwt),

First of all, get used to physical units. All physics is about quantities and their rate of change. I will write this down in a very trivial way, just to have the concept on the table (of course I do not take you for a five year old, please just bear with me :wink:). Power is the rate of change of energy, so if over one hour you constantly apply 100 Watts of power, then in total you used one Watt-hour of energy. Same as with speed and distance, if you spend one hour driving at 100 kilometers per hour, you end up driving 100 kilometers in total. For you as an engineer:

$$W = \int P \textrm{d} t \qquad \textrm{and} \qquad s = \int v \textrm{d} t$$

So yes, you have to distinguish Power and Work (energy)! OM has the ability to express this, there is unit division (km/h), unit multiplication (W · h), and everything else you need to obtain unit combinations. If you are faced with new units (e.g., CO2 equivalents per amount of energy), OM can help :sweat_smile:!

With the speed vs distance example, you probably noticed that quantities can have direction or not. The English language distinguishes speed (a value) and velocity (a vector). As you pointed out, quantities can be positive and negative. However, what you mean by that is not part of the unit. Let's say you sell energy from your PV installation to your neighbour. Some of that he uses in his dishwasher, some other he stores in his home battery. Later that day he sells you energy from his battery back and you run your washing machine on that. Which direction is positive here?

So much for the trivial part. Here is the hard part. In the InterConnect project (your colleagues participated), we developed a Data Point ontology that aimed to capture all those distinctions (see https://gitlab.inesctec.pt/groups/interconnect-public/-/wikis/home#interconnect-ontology), but still fell short. Maybe have a look at https://gitlab.inesctec.pt/interconnect-public/ontology/-/wikis/ic-data, specifically at the concept Usage (which should have an ontology of its own!).

There is still a lot of work to do here... :weary: Let us know about your progress.

Cheers

[GeorgJung]

Hi Barry (@bnouwt), Checking back the InterConnect ontologies it is now blatantly obvious to me that putting this Usage concept into the ic-data ontology was essentially a blunder. Instead, there should be some dedicated ontology covering the various terms in energy infrastructure such as feed-in, offtake, production, consumption, charging, discharging, overvoltage, undervoltage, overcurrent, capacity, charge, state of charge, and so on (might not even be all that many concepts,,,). Many are anti-pairs, a negative offtake is obviously a positive feed-in, others are connected, the state of charge is the charge divided by the capacity, and so on. Maybe that is what you should look into, not adding this to OM but creating such an ontology that then uses OM.

Oh, and let me know if you do! If I find a project to squeeze this in and thus have some time addressing it, I will let you know too.

Cheers