Air leakage type "unit total"

[joseph-robertson]

Pull Request Description

This PR fixed a bug that resulted in the modeled ACH50 infiltration values of the units being larger than the sampled ACH50 for the building. The update is to change options_lookup and ResStockArguments to use unit total air leakage type instead of the current unit exterior only type w/ infiltration adjustment approach. This change provides the infiltration value sampled as the total infiltration from the exterior and other neighboring units. The infiltration to the exterior is then calculated as a percentage of the exterior surfaces. This change will reduce the modeled infiltration for MF units. The modeled infiltration to the exterior (build_existing_model.air_leakage_to_outside_ach_50) is now reported in the results.

Updates

Remove the "Infiltration adjustment for SFA/MF units" block of code from ResStockArguments.
Update BuildExistingModel to register adjusted total infiltration ACH50 value.
Some related plots in: #728

Some clarification (need documentation update):

• Sampled ACH 50 is the total unit leakage from outside and adjacent units
  • Add/update "Assumptions" in the Infiltration.tsv file?
• The output leakage is the simulated leakage to the outside
  • We've updated data/dictionary/inputs.csv with an build_existing_model.air_leakage_to_outside_ach_50 entry along with a description: "Total infiltration to the dwelling unit adjusted by ratio of exterior envelope surface area to total envelope surface area."

Checklist

Not all may apply:

• Tests (and test files) have been updated
• Documentation has been updated
  • If related to resstock-estimation, checklist includes data dictionary, source report, options saturation, options_lookup.
• Add to the changelog_dev.rst file
• openstudio tasks.rb update_measures has been run
• No unexpected regression test changes on CI (checked comparison artifacts)

[joseph-robertson]

Next steps here are, in order:

• look at energy impacts (especially heating, cooling)
• compare our distributions to ResDB
• replace the new cost multiplier approach with, e.g,, build_existing_model.air_infiltration_leakage_to_outside_xxx

[joseph-robertson]

Questions from meeting:

• How do we compare new ResStock distributions to what we have available on ResDB? Qualitative comparison?
• Convert from Normalized leakage (NL) to ACH50?

[afontani]

Questions from meeting:

• How do we compare new ResStock distributions to what we have available on ResDB? Qualitative comparison?
• Convert from Normalized leakage (NL) to ACH50?

Maybe compare the relative multipliers for the different building types (see the last two columns)

We are generally the same as ResDB for the whole stock.

[joseph-robertson]

Meeting notes:

• Use the data we generated the ACH50 plots from to get percentiles. Compare these to the ResDB SFD numbers times multipliers in the tables above.
• This PR is a bugfix PR. There may be a follow-up PR to adjust distributions to better match ResDB.

[joseph-robertson]

Infiltration sources	5th	50th	95th	Segment
ResDB	4	12	35	Mostly single-family detached
ResDB	5.2	15.6	45.5	Single-Family Attached Homes
ResDB	2.8	8.4	24.5	Multi-Family Homes
ResDB	6.16	18.48	53.9	Manufactured Homes

Infiltration sources	5th	50th	95th	Segment
ResStock	3	15	40	U.S. residential stock
ResStock	4	15	40	Single-Family Detached
ResStock	3.436187	10.19865	26.85008	Single-Family Attached
ResStock	1.0734	8.1402	26.8074	Multi-Family

[ekpresent]

Sorry, this PR caught my attention and I have a couple naive questions.

1. Are there revised plots available like the ones @joseph-robertson posted on June 13?
2. Was an energy results comparison ever completed? (sorry if I just don't know where to look for it) - I'd be interested in segmentation by at least building type and climate zone.

[joseph-robertson]

Plan is to:

• run 10K downselected to only SFA/MF on develop
• run same datapoints using this feature branch
• plots of heating/cooling, sliced by building type and climate zone

[joseph-robertson]

@ekpresent Here are before (i.e., base) and after (i.e., feature) average % Error results for 10K datapoints.

Across the whole set:

	base	feature	% Error
energy_use_total_m_btu	101.5169	99.96975	-1.52407
fuel_use_electricity_total_m_btu	39.426	39.08133	-0.87422
fuel_use_natural_gas_total_m_btu	54.83362	53.72353	-2.02448
fuel_use_fuel_oil_total_m_btu	5.812796	5.720423	-1.58914
fuel_use_propane_total_m_btu	1.229316	1.229267	-0.00395
end_use_electricity_heating_m_btu	5.640149	5.365818	-4.86389
end_use_electricity_heating_fans_pumps_m_btu	1.216259	1.197097	-1.57551
end_use_electricity_heating_heat_pump_backup_m_btu	0.482422	0.472269	-2.10459
end_use_electricity_heating_heat_pump_backup_fans_pumps_m_btu	0.014769	0.014382	-2.61704
end_use_electricity_cooling_m_btu	8.493371	8.44879	-0.52489
end_use_electricity_cooling_fans_pumps_m_btu	1.411581	1.414026	0.173167
end_use_natural_gas_heating_m_btu	45.45434	44.34519	-2.44014
end_use_natural_gas_heating_heat_pump_backup_m_btu	0	0
end_use_fuel_oil_heating_m_btu	5.439992	5.347677	-1.69696
end_use_fuel_oil_heating_heat_pump_backup_m_btu	0	0
end_use_propane_heating_m_btu	0.963625	0.963625	0
end_use_propane_heating_heat_pump_backup_m_btu	0	0

Segmented by:

• geometry_building_type_recs: averages_geometry_building_type_recs.csv
• ashrae_iecc_climate_zone_2004: averages_ashrae_iecc_climate_zone_2004.csv

[rajeee]

These enduses are going up for some reason. Might be worth finding out why.

	geometry_building_type_recs	base	feature	% Error
end_use_electricity_cooling_fans_pumps_m_btu	Single-Family Attached	0.75801682	0.78837944	4.00553338
end_use_electricity_cooling_fans_pumps_m_btu	Multi-Family with 5+ Units	0.46667142	0.47255323	1.26037515
end_use_electricity_cooling_fans_pumps_m_btu	Multi-Family with 2 - 4 Units	0.23117381	0.23384424	1.15516063
end_use_electricity_cooling_fans_pumps_m_btu	All	1.4115813	1.4140257	0.1731675

[joseph-robertson]

These enduses are going up for some reason. Might be worth finding out why.

  geometry_building_type_recs base feature % Error
end_use_electricity_cooling_fans_pumps_m_btu Single-Family Attached 0.75801682 0.78837944 4.00553338
end_use_electricity_cooling_fans_pumps_m_btu Multi-Family with 5+ Units 0.46667142 0.47255323 1.26037515
end_use_electricity_cooling_fans_pumps_m_btu Multi-Family with 2 - 4 Units 0.23117381 0.23384424 1.15516063
end_use_electricity_cooling_fans_pumps_m_btu All 1.4115813 1.4140257 0.1731675

For example, ID=21. Modeled ACH50 for this MF unit changes from 8.09 to 0.59. HeatingCapacity (furnace) changes from 6188 to 1860. CoolingCapacity (mini-split air conditioner) changes from 8097 to 6066. Indeed cooling increases slightly. We'll need to look into this. @shorowit do you have any insight here?

[joseph-robertson]

These enduses are going up for some reason. Might be worth finding out why.
  geometry_building_type_recs base feature % Error
end_use_electricity_cooling_fans_pumps_m_btu Single-Family Attached 0.75801682 0.78837944 4.00553338
end_use_electricity_cooling_fans_pumps_m_btu Multi-Family with 5+ Units 0.46667142 0.47255323 1.26037515
end_use_electricity_cooling_fans_pumps_m_btu Multi-Family with 2 - 4 Units 0.23117381 0.23384424 1.15516063
end_use_electricity_cooling_fans_pumps_m_btu All 1.4115813 1.4140257 0.1731675

For example, ID=21. Modeled ACH50 for this MF unit changes from 8.09 to 0.59. HeatingCapacity (furnace) changes from 6188 to 1860. CoolingCapacity (mini-split air conditioner) changes from 8097 to 6066. Indeed cooling increases slightly. We'll need to look into this. @shorowit do you have any insight here?

This could be due to a high amount of internal gains -- infiltration was previously aiding the cooling. Should check number of occupants, plug load multiplier.

[joseph-robertson]

These enduses are going up for some reason. Might be worth finding out why.
  geometry_building_type_recs base feature % Error
end_use_electricity_cooling_fans_pumps_m_btu Single-Family Attached 0.75801682 0.78837944 4.00553338
end_use_electricity_cooling_fans_pumps_m_btu Multi-Family with 5+ Units 0.46667142 0.47255323 1.26037515
end_use_electricity_cooling_fans_pumps_m_btu Multi-Family with 2 - 4 Units 0.23117381 0.23384424 1.15516063
end_use_electricity_cooling_fans_pumps_m_btu All 1.4115813 1.4140257 0.1731675

For example, ID=21. Modeled ACH50 for this MF unit changes from 8.09 to 0.59. HeatingCapacity (furnace) changes from 6188 to 1860. CoolingCapacity (mini-split air conditioner) changes from 8097 to 6066. Indeed cooling increases slightly. We'll need to look into this. @shorowit do you have any insight here?

This could be due to a high amount of internal gains -- infiltration was previously aiding the cooling. Should check number of occupants, plug load multiplier.

ID=21 is:

• unit in a 183-unit apartment building
• 2 bed, 1138 cfa
• MD, North Bethesda
• 2 occupants
• electric furnace, shared fan coil
• plug load multiplier 1.06

[afontani]

Tony to update description and add a description in the assumptions about how we are modeling infiltration for multi-family units.

Description: Total infiltration to the dwelling unit adjusted by ratio of exterior envelope surface area to total envelope surface area