Entering/Leaving Chilled_Water_Temperature definitions.

[sappling]

Currently these two classes are defined as:
Leaving_Chilled_Water_Temperature_Sensor : Measures the temperature of chilled water that is supplied from a chiller
Entering_Chilled_Water_Temperature_Sensor : Measures the temperature of chilled water that is enteringed to a cooling tower

I have several concerns and would like to get consensus before creating a pull request.

1. Entering_Chilled_Water_Temperature_Sensor is not the temperature of the water entering the cooling tower. That would be condenser water.
2. I don't know that these definitions should refer to the chiller. I think that perhaps they should be like Entering_Chilled_Water_Flow_Sensor that doesn't refer to the chiller, but just refers to "chilled entering water" (although I would prefer "entering chilled water").
3. How would these be used outside of a chiller? If a water cooled AHU has a point measuring the temperature of it's incoming chilled water, it would seem to be the Entering_Chilled_Water_Temperature_Sensor, but by the above definitions, it should probably be the Leaving one since that is defined as what is supplied by the chiller. I think that having the chiller as part of the definition is confusing.
4. Lastly - I don't think that there is a clear way of distinguishing the chilled water leaving a specific chiller from the chilled water leaving the whole chiller plant. This was brought up in Issue 628, but doesn't seem to be resolved.

I am proposing:
Leaving_Chilled_Water_Temperature_Sensor : Measures the temperature of chilled water that is leaving an equipment
Entering_Chilled_Water_Temperature_Sensor : Measures the temperature of chilled water that is entering an equipment

To me this would mean that the a point in an AHU measuring the temperature of it's incoming chilled water would be an Entering_Chilled_Water_Temperature_Sensor.

I'm not clear how to model the chilled water for the whole chiller plant. Would this be a Leaving_Chilled_Water_Temperature_Sensor on a Chilled_Water_System?

[gtfierro]

Hi @sappling ! There's an older issue that raises some of these issues, too: #677 . I think we need consensus on what Entering/Leaving/Supply/Return mean and how they relate to the structure of the system. That PR proposes using supply/return on the load side of a water system, and entering/leaving on the plant side.

I think there is consensus on preferring "entering chilled water" over "chilled entering water". Can you tell we made some of these classes using find-replace? :). I also agree they shouldn't refer to the chiller or any equipment in the class name.

How would these be used outside of a chiller? If a water cooled AHU has a point measuring the temperature of it's incoming chilled water, it would seem to be the Entering_Chilled_Water_Temperature_Sensor, but by the above definitions, it should probably be the Leaving one since that is defined as what is supplied by the chiller. I think that having the chiller as part of the definition is confusing.

Per #677, if we decide to use that proposed model, the AHU would have a Supply Chilled Water Temp Sensor for the water it is receiving from the chiller system.

Lastly - I don't think that there is a clear way of distinguishing the chilled water leaving a specific chiller from the chilled water leaving the whole chiller plant. Would this be a Leaving_Chilled_Water_Temperature_Sensor on a Chilled_Water_System?

Yep! You would just associate the leaving chilled water sensor with the system, and other sensors with the constituent chillers.

<chilled-water-system> brick:hasPoint :chws-leaving-chilled-water-temp-sensor ;
   brick:hasPart <chiller1>, <chiller2> .

<chiller1> brick:hasPoint :leaving-chilled-water-temp-sensor1 .
<chiller2> brick:hasPoint :leaving-chilled-water-temp-sensor2 .

[gtfierro]

I'm commenting the Project Haystack model here (https://project-haystack.org/doc/docHaystack/chilled-water-plant.svg) so I can refer to it when I get to working on this. This suggests we want the substance (chiller vs condenser water) to encode which loop we are talking about

[sappling]

Hmmm. That picture made me think more about the cooling tower side. While supply/return makes sense to me on the chiller, I'm less sure of naming on the condenser water side. I wouldn't think of either the chiller or cooling tower supplying condenser water. I think both of them have entering/leaving condenser water temperature points. If it's not obvious for a particular piece of equipment what is supply/return vs entering/leaving, then perhaps we shouldn't differentiate.

[gtfierro]

Hey @sappling I put together a Brick model of the cooling tower+chiller plant in that Haystack diagram: https://github.com/BrickSchema/Brick/blob/e846b8c5e1a29de7f5eda391ddca6ade78336df1/examples/chiller-system/chiller-plant-haystack.ttl . We can see how entering/leaving relate to the equipment they are on: "entering" the cooling tower from the chiller vs "entering" the chiller from the cooling tower. We can also see the use of supply/return to indicate the load-side of the chiller. What do you think of the model?

[sappling]

@gtfierro, after looking at your sample, the diagram above, and your proposed changes in fix-issue-743, I'm still not sure that distinguishing between entering/leaving and supply/return is helpful for defining this formally. I agree that there is common industry usage of these terms, but I don't know that the definitions you provided in the fix for 743 are consistent across all uses.

You are distinguishing Entering_Water from Returning_Water by defining it as

Water that is entering a piece of equipment within a water system; contrasts with returning water which is specifically water coming back from the load side of the system.

Is the "load side of the system" clear on the condenser water side? Is the chiller the load for condenser water?

I think that project haystack just ended up with just using entering and leaving. I think adding supply/return back will actually be confusing and you should keep using just entering/leaving like haystack does, but fix some of the current broken entering/leaving definitions.