Skip to content

Commit

Permalink
Different approach to calculating EndUse Balance is introduced (#324)
Browse files Browse the repository at this point in the history 
* Uses the "Zone Predicted Sensible Load to Setpoint Heat Transfer Rate" to determine if zone is in cooling or in heating

* Cleans up required outputs

* Uses Predicted load to Temp to figure out if coolinf or heating is happening

* remove mech_ventilation
  • Loading branch information
Samuel Letellier-Duchesne authored Jun 14, 2022
1 parent a1f1a72 commit d0aed03
Show file tree
Hide file tree
Showing 3 changed files with 111 additions and 124 deletions.
225 changes: 107 additions & 118 deletions archetypal/idfclass/end_use_balance.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,7 @@


class EndUseBalance:
HVAC_MODE = ("Zone Predicted Sensible Load to Setpoint Heat Transfer Rate",)
HVAC_INPUT_SENSIBLE = ( # not multiplied by zone or group multipliers
"Zone Air Heat Balance System Air Transfer Rate",
"Zone Air Heat Balance System Convective Heat Gain Rate",
Expand All @@ -34,52 +35,36 @@ class EndUseBalance:
"Zone Hot Water Equipment Convective Heating Energy",
"Zone Other Equipment Convective Heating Energy",
)
PEOPLE_GAIN = ("Zone People Sensible Heating Energy",) # checked, Todo: +latent
PEOPLE_GAIN = ("Zone People Total Heating Energy",) # checked
SOLAR_GAIN = ("Zone Windows Total Transmitted Solar Radiation Energy",) # checked
INFIL_GAIN = (
"Zone Infiltration Sensible Heat Gain Energy", # checked
# "Zone Infiltration Latent Heat Gain Energy",
"AFN Zone Infiltration Sensible Heat Gain Energy",
# "AFN Zone Infiltration Latent Heat Gain Energy",
"Zone Infiltration Total Heat Gain Energy", # checked
"AFN Zone Infiltration Total Heat Gain Energy",
)
INFIL_LOSS = (
"Zone Infiltration Sensible Heat Loss Energy", # checked
# "Zone Infiltration Latent Heat Loss Energy",
"AFN Zone Infiltration Sensible Heat Loss Energy",
# "AFN Zone Infiltration Latent Heat Loss Energy",
"Zone Infiltration Total Heat Loss Energy", # checked
"AFN Zone Infiltration Total Heat Loss Energy",
)
VENTILATION_LOSS = ("Zone Air System Sensible Heating Energy",)
VENTILATION_GAIN = ("Zone Air System Sensible Cooling Energy",)
VENTILATION_LOSS = ("Zone Air System Total Heating Energy",)
VENTILATION_GAIN = ("Zone Air System Total Cooling Energy",)
NAT_VENT_GAIN = (
# "Zone Ventilation Total Heat Gain Energy",
"Zone Ventilation Sensible Heat Gain Energy",
# "Zone Ventilation Latent Heat Gain Energy",
"AFN Zone Ventilation Sensible Heat Gain Energy",
# "AFN Zone Ventilation Latent Heat Gain Energy",
"Zone Ventilation Total Heat Gain Energy",
"AFN Zone Ventilation Total Heat Gain Energy",
)
NAT_VENT_LOSS = (
# "Zone Ventilation Total Heat Loss Energy",
"Zone Ventilation Sensible Heat Loss Energy",
# "Zone Ventilation Latent Heat Loss Energy",
"AFN Zone Ventilation Sensible Heat Loss Energy",
# "AFN Zone Ventilation Latent Heat Loss Energy",
)
MECHANICAL_VENT_LOSS = (
"Zone Mechanical Ventilation No Load Heat Removal Energy",
"Zone Mechanical Ventilation Heating Load Increase Energy",
"Zone Mechanical Ventilation Cooling Load Decrease Energy",
)
MECHANICAL_VENT_GAIN = (
"Zone Mechanical Ventilation No Load Heat Addition Energy",
"Zone Mechanical Ventilation Heating Load Decrease Energy",
"Zone Mechanical Ventilation Cooling Load Increase Energy",
"Zone Ventilation Total Heat Loss Energy",
"AFN Zone Ventilation Total Heat Loss Energy",
)
OPAQUE_ENERGY_FLOW = ("Surface Outside Face Conduction Heat Transfer Energy",)
OPAQUE_ENERGY_STORAGE = ("Surface Heat Storage Energy",)
WINDOW_LOSS = ("Zone Windows Total Heat Loss Energy",) # checked
WINDOW_GAIN = ("Zone Windows Total Heat Gain Energy",) # checked
HEAT_RECOVERY_LOSS = ("Heat Exchanger Total Cooling Energy",)
HEAT_RECOVERY_GAIN = ("Heat Exchanger Total Heating Energy",)
HRV_LOSS = ("Heat Exchanger Total Cooling Energy",)
HRV_GAIN = ("Heat Exchanger Total Heating Energy",)
AIR_SYSTEM = (
"Air System Heating Coil Total Heating Energy",
"Air System Cooling Coil Total Cooling Energy",
)

def __init__(
self,
Expand All @@ -100,6 +85,7 @@ def __init__(
opaque_storage,
window_flow,
heat_recovery,
air_system,
is_cooling,
is_heating,
units="J",
Expand All @@ -122,6 +108,7 @@ def __init__(
self.opaque_storage = opaque_storage
self.window_flow = window_flow
self.heat_recovery = heat_recovery
self.air_system = air_system
self.units = units
self.use_all_solar = use_all_solar
self.is_cooling = is_cooling
Expand Down Expand Up @@ -168,15 +155,15 @@ def from_sql_file(
axis=1,
verify_integrity=True,
)

rolling_sign = cls.get_rolling_sign_change(_hvac_input)
mode = sql.timeseries_by_name(cls.HVAC_MODE) # positive = Heating
rolling_sign = cls.get_rolling_sign_change(mode).fillna(0)

# Create both heating and cooling masks
is_heating = rolling_sign > 0
is_cooling = rolling_sign < 0
is_heating = rolling_sign.droplevel(["IndexGroup", "Name"], axis=1) == 1
is_cooling = rolling_sign.droplevel(["IndexGroup", "Name"], axis=1) == -1

heating = _hvac_input[is_heating].fillna(0)
cooling = _hvac_input[is_cooling].fillna(0)
heating = _hvac_input.mul(is_heating, level="KeyValue", axis=1)
cooling = _hvac_input.mul(is_cooling, level="KeyValue", axis=1)

lighting = sql.timeseries_by_name(cls.LIGHTING).to_units(units)
zone_multipliers = sql.zone_info.set_index("ZoneName")["Multiplier"].rename(
Expand Down Expand Up @@ -204,23 +191,10 @@ def from_sql_file(
nat_vent_gain = cls.apply_multipliers(nat_vent_gain, zone_multipliers)
nat_vent_loss = sql.timeseries_by_name(cls.NAT_VENT_LOSS).to_units(units)
nat_vent_loss = cls.apply_multipliers(nat_vent_loss, zone_multipliers)
mech_vent_gain = sql.timeseries_by_name(cls.MECHANICAL_VENT_GAIN).to_units(
units
)
mech_vent_gain = cls.apply_multipliers(mech_vent_gain, zone_multipliers)
mech_vent_loss = sql.timeseries_by_name(cls.MECHANICAL_VENT_LOSS).to_units(
units
)
mech_vent_loss = cls.apply_multipliers(mech_vent_loss, zone_multipliers)
heat_recovery_loss = sql.timeseries_by_name(cls.HEAT_RECOVERY_LOSS).to_units(
units
)
heat_recovery_gain = sql.timeseries_by_name(cls.HEAT_RECOVERY_GAIN).to_units(
units
)
heat_recovery = cls.subtract_loss_from_gain(
heat_recovery_gain, heat_recovery_loss, level="KeyValue"
)
hrv_loss = sql.timeseries_by_name(cls.HRV_LOSS).to_units(units)
hrv_gain = sql.timeseries_by_name(cls.HRV_GAIN).to_units(units)
hrv = cls.subtract_loss_from_gain(hrv_gain, hrv_loss, level="KeyValue")
air_system = sql.timeseries_by_name(cls.AIR_SYSTEM).to_units(units)

# subtract losses from gains
infiltration = None
Expand All @@ -230,10 +204,6 @@ def from_sql_file(
infiltration = cls.subtract_loss_from_gain(
infil_gain, infil_loss, level="Name"
)
if not any((vent_gain.empty, vent_loss.empty, cooling.empty, heating.empty)):
mech_vent = cls.subtract_loss_from_gain(
mech_vent_gain, mech_vent_loss, level="Name"
)
if nat_vent_gain.shape == nat_vent_loss.shape:
nat_vent = cls.subtract_loss_from_gain(
nat_vent_gain, nat_vent_loss, level="Name"
Expand Down Expand Up @@ -296,7 +266,8 @@ def from_sql_file(
opaque_flow,
opaque_storage,
window_flow,
heat_recovery,
hrv,
air_system,
is_cooling,
is_heating,
units,
Expand Down Expand Up @@ -492,9 +463,7 @@ def subtract_vent_from_system(cls, system, vent, level="Key_Name"):
index=system.index,
)

def separate_gains_and_losses(
self, component, level="Key_Name", stack_on_level=None
) -> EnergyDataFrame:
def separate_gains_and_losses(self, component, level="Key_Name") -> EnergyDataFrame:
"""Separate gains from losses when cooling and heating occurs for the component.
Args:
Expand All @@ -509,38 +478,31 @@ def separate_gains_and_losses(
), f"{component} is not a valid attribute of EndUseBalance."
component_df = getattr(self, component)
assert not component_df.empty, "Expected a component that is not empty."
if isinstance(level, str):
level = [level]
print(component)

# mask when cooling occurs in zone (negative values)
mask = (self.is_cooling.stack("KeyValue") == True).any(axis=1)

# get the dataframe using the attribute name, summarize by `level` and stack so that a Series is returned.
stacked = getattr(self, component).sum(level=level, axis=1).stack(level[0])
c_df = getattr(self, component)

# concatenate the masked values with keys to easily create a MultiIndex when unstacking
# concatenate the Periods
inter = pd.concat(
[
stacked[mask].reindex(stacked.index),
stacked[~mask].reindex(stacked.index),
c_df.mul(self.is_cooling.rename_axis(level, axis=1), level=level),
c_df.mul(self.is_heating.rename_axis(level, axis=1), level=level),
],
keys=["Cooling Periods", "Heating Periods"],
names=["Period"]
+ stacked.index.names, # prepend the new key name to the existing index names.
names=["Period"],
)

# mask when values are positive (gain)
positive_mask = inter >= 0

# concatenate the masked values with keys to easily create a MultiIndex when unstacking
# concatenate the Gain/Loss
final = pd.concat(
[
inter[positive_mask].reindex(inter.index),
inter[~positive_mask].reindex(inter.index),
],
keys=["Heat Gain", "Heat Loss"],
names=["Gain/Loss"] + inter.index.names,
names=["Gain/Loss"],
).unstack(["Period", "Gain/Loss"])
final.sort_index(axis=1, inplace=True)
return final
Expand All @@ -560,40 +522,24 @@ def to_df(self, separate_gains_and_losses=False, level="KeyValue"):
"infiltration",
"window_energy_flow",
"nat_vent",
"mech_vent",
]:
if not getattr(self, component).empty:
summary_by_component[component] = (
self.separate_gains_and_losses(
component,
level=level,
)
.unstack(level)
.reorder_levels([level, "Period", "Gain/Loss"], axis=1)
.groupby(level=["KeyValue", "Period", "Gain/Loss"], axis=1)
.sum()
.sort_index(axis=1)
)
for (surface_type), data in (
self.separate_gains_and_losses(
"opaque_flow", ["Zone_Name", "Surface_Type"]
)
.unstack("Zone_Name")
.groupby(level=["Surface_Type"], axis=1)
):
for (surface_type), data in self.separate_gains_and_losses(
"opaque_flow", level="Zone_Name"
).groupby(level=["Surface_Type"], axis=1):
summary_by_component[surface_type] = data.sum(
level=["Zone_Name", "Period", "Gain/Loss"], axis=1
).sort_index(axis=1)

# for (surface_type), data in (
# self.separate_gains_and_losses(
# "opaque_storage", ["Zone_Name", "Surface_Type"]
# )
# .unstack("Zone_Name")
# .groupby(level=["Surface_Type"], axis=1)
# ):
# summary_by_component[surface_type + " Storage"] = data.sum(
# level=["Zone_Name", "Period", "Gain/Loss"], axis=1
# ).sort_index(axis=1)

else:
summary_by_component = {}
for component in [
Expand All @@ -606,20 +552,40 @@ def to_df(self, separate_gains_and_losses=False, level="KeyValue"):
"infiltration",
"window_energy_flow",
"nat_vent",
"mech_vent",
]:
component_df = getattr(self, component)
if not component_df.empty:
summary_by_component[component] = component_df.sum(
level=level, axis=1
).sort_index(axis=1)
for (zone_name, surface_type), data in self.face_energy_flow.groupby(
for (zone_name, surface_type), data in self.opaque_flow.groupby(
level=["Zone_Name", "Surface_Type"], axis=1
):
summary_by_component[surface_type] = data.sum(
level="Zone_Name", axis=1
).sort_index(axis=1)
levels = ["Component", "Zone_Name"]

# Add contribution of heating/cooling outside air, if any
if not self.air_system.empty:
summary_by_component["air_system_heating"] = (
self.air_system.xs(
"Air System Heating Coil Total Heating Energy", level="Name", axis=1
)
.assign(**{"Period": "Heating Periods", "Gain/Loss": "Heat Loss"})
.set_index(["Period", "Gain/Loss"], append=True)
.unstack(["Period", "Gain/Loss"])
.droplevel("IndexGroup", axis=1)
)
summary_by_component["air_system_cooling"] = (
self.air_system.xs(
"Air System Cooling Coil Total Cooling Energy", level="Name", axis=1
)
.assign(**{"Period": "Cooling Periods", "Gain/Loss": "Heat Gain"})
.set_index(["Period", "Gain/Loss"], append=True)
.unstack(["Period", "Gain/Loss"])
.droplevel("IndexGroup", axis=1)
)
return pd.concat(
summary_by_component, axis=1, verify_integrity=True, names=levels
)
Expand Down Expand Up @@ -693,6 +659,9 @@ def to_sankey(self, path_or_buf):
"interior_equipment": "Equipment",
"window_energy_flow": "Windows",
"Wall": "Walls",
"air_system_heating": "OA Heating",
"air_system_cooling": "OA Cooling",
"cooling": "Cooling",
},
inplace=True,
)
Expand Down Expand Up @@ -752,9 +721,9 @@ def to_sankey(self, path_or_buf):
)

system_input = (
system_input
# .replace({0: np.NaN})
.dropna(how="all").dropna(how="all", axis=1)
system_input.replace({0: np.NaN})
.dropna(how="all")
.dropna(how="all", axis=1)
)
system_input.rename_axis("source", axis=1, inplace=True)
system_input.rename_axis("target", axis=0, inplace=True)
Expand Down Expand Up @@ -789,21 +758,41 @@ def to_sankey(self, path_or_buf):
link_cooling_system_to_gains,
) = self._sankey_cooling(cooling_load, load_type="cooling")

return (
pd.DataFrame(
system_input_data
+ link_heating_system_to_gains
+ heating_energy_to_heating_system
+ heating_load_source_data
+ heating_load_target_data
+ cooling_energy_to_heating_system
+ cooling_load_source_data
+ cooling_load_target_data
+ link_cooling_system_to_gains
)
# .div(floor_area)
.to_csv(path_or_buf, index=False)
flows = pd.DataFrame(
system_input_data
+ heating_energy_to_heating_system
+ heating_load_source_data
+ heating_load_target_data
+ cooling_energy_to_heating_system
+ cooling_load_source_data
+ cooling_load_target_data
)
# Fix the HVAC difference
diff = (
flows.loc[flows.source == "Heating Load", "value"].sum()
- flows.loc[flows.target == "Heating Load", "value"].sum()
)
flows.loc[flows.source == "Heating System", "value"] = (
flows.loc[flows.source == "Heating System", "value"] + diff
)

diff = (
flows.loc[flows.source == "Cooling Load", "value"].sum()
- flows.loc[flows.target == "Cooling Load", "value"].sum()
)
flows.loc[flows.source == "Cooling System", "value"] = (
flows.loc[flows.source == "Cooling System", "value"] + diff
)

# fix names
flows.replace({"OA Heating Heat Losses": "OA Heating"}, inplace=True)

# TO EUI
flows["value"] = flows["value"] / floor_area
links = pd.DataFrame(
link_heating_system_to_gains + link_cooling_system_to_gains
)
return pd.concat([flows, links]).to_csv(path_or_buf, index=False)

def _sankey_heating(self, load, load_type="heating"):
assert load_type in ["heating", "cooling"]
Expand Down
Loading

0 comments on commit d0aed03

Please sign in to comment.