QUDT Questions for Brick 2.0 Point Attributes

[gtfierro]

Context: we are adding attributes to Brick point classes to support better queries and to ensure greater consistency in how points are modeled. See the following for more context:

• https://github.com/BrickSchema/Brick2-specification-drafting/tree/main/point-attributes
• Missing hasQuantity and hasSubstance properties for some Point classes #713
• Consistency of hasQuantity/hasSubstance on Point subclasses and Point instances #696

Missing or Incomplete Quantity Kinds:
- Zenith (angle) is missing from QUDT. Can we add?

1. Electric Energy
   • QUDT has ElectricPower but not ElectricEnergy as a distinct quantity kind.
   • Is there a reason QUDT doesn't have ElectricEnergy? Should we add it?
2. Thermal Power
   • QUDT has ThermalEnergy but not ThermalPower.
   • We encounter points measuring thermal power (e.g., heating/cooling power output).
   • Can we add ThermalPower to QUDT?
3. Reactive Energy
   • We encounter points with units of kvarh (kilovolt-ampere reactive hours).
   • I couldn't find a reactive energy QK. Is this in scope for QUDT?
4. Current/Voltage Imbalance
   • We want to represent current/voltage imbalance (percentage deviation from average current or voltage across phases).
   • I'm 90% sure we can just use PERCENT for this on units, but is there a good QK to use?

Questions on Units / Missing Units:

• We would like to add pCi/L (picocuries per liter) for radon activity concentration points
• Grains of Moisture
  • air grains sensors use "grains of water vapor per pound of dry air" as their unit.
  • QUDT has a GRAIN unit (a unit of mass), but the relevant physical quantity for Air Grains is a mass ratio (mass of water vapor / mass of dry air).
  • Does this merit creation of a new unit in QUDT?

Misc:

• how to handle Window Tint? We have Window_Tint_Command and Window_Tint_Status points in Brick
• unclear if there is a standard unit for window tint (% UV blocked? lumens attenuated?). Sometimes there are unusual units for these ("Oktas" for cloud coverage was a new one for me)

@steveraysteveray @michaelpoplawskipnnl Here are the questions that arose from @ektrah and I investigating the Brick point types. I imagine these changes would also be helpful for 223P as well! Please let me know if I can provide any more context on any of the above

[ektrah]

Discussion with @steveraysteveray @ralphtq @mbmaja:

• ElectricEnergy, ThermalPower, ReactiveEnergy, CurrentImbalance, and VoltageImbalance make sense to add at first glance; @ektrah will submit a PR to the QUDT repository
• "solar azimuth", "solar zenith", and "wind direction": are solar and wind "substances" or should "solar" and "wind" be part of the QuantityKind name? what's the right quantity kind for "direction"?
• @steveraysteveray will look at the other questions (units and misc)
• Also discussed: correct modeling of differences, such as temperature differences and pressure differences
  • for temperatures, it's currently unclear whether to use hasQuantityKind Temperature and isDeltaQuantity true or hasQuantityKind TemperatureDifference and isDeltaQuantity false
  • for pressure, there is only hasQuantityKind Pressure and isDeltaQuantity true but no PressureDifference
  • adding a ___Difference for every existing quantity kind would add a lot of noise to QUDT
  • however, isDeltaQuantity is intended for quantities, i.e., not suitable for cases where no specific numeric value is given

[steveraysteveray]

To address the bullets above:

• First bullet: Those all look like valid QuantityKind submissions
• Second bullet: The solar ones sound good. Presumably, the normal applicable unit would be unit:DEG. That unit definition should include a 'qudt:hasQuantityKind' reference to your new quantity kinds
• Second bullet still: WindDirection should be documented as the angle from true north, or magnetic north, (you should make clear which one), in the clockwise direction as per convention. Same treatment from unit:DEG.
• Still discussing the final bullet about differences...

Regarding @gtfierro's original issue above:

• picocuries per liter would be unit:PicoCI-PER-L. Sounds good.
• If it really is a thing, you could add GRAIN_AIR, defined as grains of moisture per pound of dry air, and figure out the conversion multiplier to KiloGM-PER-KiloGM. Presumably the conversion multiplier would be the same magnitude as converting to pound per pound. Make sure the dimension vector is A0E0L0I0M0H0T0D1, but you could include the qudt:qkdvNumerator and qudt:qkdvDenominator triples to identify it as a mass ratio (A0E0L0I0M1H0T0D0).
• I don't know about window tints, but presumably the unit would be a percent. You would need a quantity kind that would define whether 100% means totally opaque or totally transparent (much like we defined quantitykind:OpeningRatio for dampers and valves).

That's all I have for now. Sorry for the delay in responding, @gtfierro.

[gtfierro]

Don't worry about the delay! This is a very helpful response. I'll start assembling the changes and will get back in touch if I have any questions.