Heat Pump
The epl.HeatPump asset is suitable for modelling electric heat pumps.
It uses electricity to convert low temperature heat to high temperature heat.
Assumptions
When using epl.HeatPump.optimize, the alternative to the heat pump is generating high temperature heat from a gas boiler. Under the hood of epl.HeatPump.optimize, a epl.Boiler asset is used to supply the balance of high temperature heat demand of the site.
The gas price is important as the alternative to using a heat pump to supply the high_temperature_load_mwh is using a natural gas boiler.
In order for the heat pump to work, it needs to have both a source of low temperature heat and a sink of high temperature heat.
The high_temperature_load_mwh is the amount of heat consumed by the site, and low_temperature_generation_mwh is the amount of available low temperature heat.
Use
import energypylinear as epl
asset = epl.HeatPump(
electric_power_mw=1.0,
cop=2,
gas_prices=20,
electricity_prices=[100, -100],
high_temperature_load_mwh=3.0,
low_temperature_generation_mwh=3.0,
)
simulation = asset.optimize(verbose=False)
print(
simulation.results[
[
"site-electricity_prices",
"heat-pump-electric_load_mwh",
"heat-pump-low_temperature_load_mwh",
"heat-pump-high_temperature_generation_mwh",
]
]
)
assert all(
simulation.results.columns
== [
"site-import_power_mwh",
"site-export_power_mwh",
"site-electricity_prices",
"site-electricity_carbon_intensities",
"site-high_temperature_load_mwh",
"site-low_temperature_load_mwh",
"site-low_temperature_generation_mwh",
"site-gas_prices",
"site-electric_load_mwh",
"spill-electric_generation_mwh",
"spill-electric_load_mwh",
"spill-high_temperature_generation_mwh",
"spill-low_temperature_generation_mwh",
"spill-high_temperature_load_mwh",
"spill-low_temperature_load_mwh",
"spill-gas_consumption_mwh",
"boiler-high_temperature_generation_mwh",
"boiler-gas_consumption_mwh",
"valve-high_temperature_load_mwh",
"valve-low_temperature_generation_mwh",
"heat-pump-electric_load_mwh",
"heat-pump-low_temperature_load_mwh",
"heat-pump-high_temperature_generation_mwh",
"total-electric_generation_mwh",
"total-electric_load_mwh",
"total-high_temperature_generation_mwh",
"total-low_temperature_generation_mwh",
"total-high_temperature_load_mwh",
"total-low_temperature_load_mwh",
"total-gas_consumption_mwh",
"total-electric_charge_mwh",
"total-electric_discharge_mwh",
"total-spills_mwh",
"total-electric_loss_mwh",
"site-electricity_balance_mwh",
]
)
{
'n_spills': 1,
'spill_columns': 8,
'spills': {'spill-low_temperature_load_mwh': 5.0},
'event': 'warn_spills',
'timestamp': '2023-09-08T00:41:33.743617Z',
'logger':'default_logger',
'level': 'warning'
}
site-electricity_prices heat-pump-electric_load_mwh heat-pump-low_temperature_load_mwh heat-pump-high_temperature_generation_mwh
0 100 0.0 0.0 0.0
1 -100 1.0 1.0 2.0
Under the hood the heat pump asset also includes a epl.Spill, which allows dumping of excess low temperature heat, and a epl.Valve to allow high temperature heat to flow into low temperature heat.
The combination of a epl.Spill, epl.Valve and negative electricity prices can lead to the heat pump using electricity to generate high temperature heat which is then dumped as low temperature heat. For this reason the epl.HeatPump asset includes a include_valve: bool option to turn off the valve.
You could also setup an epl.Site with other assets that generate high temperature heat to explore different tradeoffs (such as a heat pump using low temperature heat from a CHP system).
Validation
A natural response when you get access to something someone else built is to wonder - does this work correctly?
This section will give you confidence in the implementation of the heat pump asset.
Price Dispatch
Let's optimize the heat pump in two intervals - the first with a high electricity price of 100 and the second with a low electricity price of -100.
Our expectation is that the heat pump will not operate in the first interval, but will operate in the second interval:
import pandas as pd
import energypylinear as epl
pd.set_option("display.max_columns", 4)
pd.set_option('display.width', 1000)
asset = epl.HeatPump(
electric_power_mw=1.0,
cop=2,
gas_prices=20,
electricity_prices=[100, -100],
high_temperature_load_mwh=3.0,
low_temperature_generation_mwh=4.0,
)
simulation = asset.optimize(verbose=4)
print(simulation.results[
[
"site-electricity_prices",
"heat-pump-electric_load_mwh",
"boiler-high_temperature_generation_mwh"
]
])
site-electricity_prices heat-pump-electric_load_mwh boiler-high_temperature_generation_mwh
0 100.0 0.0 3.0
1 -100.0 1.0 1.0
For the first interval, with an electricity price of 100, we see that:
- our heat pump has not operated (
heat-pump-electric_load_mwh=0), - our
3.0 MWhof high temperature heat demand has been generated by the gas boiler.
For the second interval, with an electricity price of -100, we see that:
- our heat pump is operating at
1.0 MWe, which means we expect2.0 MWhof high temperature heat, - only
1.0 MWhof high temperature heat demand has been generated by the gas boiler.
Heat Balance
Without a Valve
Let's first optimize a heat pump without a high temperature to low temperature valve, which stops heat from flowing from high to low temperature.
We use a negative electricity price of -100 for each interval, to force the heat pump to operate.
import energypylinear as epl
asset = epl.HeatPump(
electric_power_mw=1.0,
cop=2,
gas_prices=20,
electricity_prices=[-100, -100, -100],
high_temperature_load_mwh=[3.0, 0.5, 3.0],
low_temperature_generation_mwh=[4.0, 4.0, 0.5],
include_valve=False
)
simulation = asset.optimize(
verbose=4,
)
print(simulation.results[
[
"site-high_temperature_load_mwh",
"site-low_temperature_generation_mwh",
"spill-low_temperature_load_mwh",
"heat-pump-electric_load_mwh",
"heat-pump-high_temperature_generation_mwh",
"boiler-high_temperature_generation_mwh"
]
])
site-high_temperature_load_mwh site-low_temperature_generation_mwh spill-low_temperature_load_mwh heat-pump-electric_load_mwh heat-pump-high_temperature_generation_mwh boiler-high_temperature_generation_mwh
0 3.0 4.0 3.00 1.00 2.0 1.0
1 0.5 4.0 3.75 0.25 0.5 0.0
2 3.0 0.5 0.00 0.50 1.0 2.0
In the first interval we are unconstrained in terms of heat - our heat pump runs at the full 1.0 MWe load.
In the second interval we have a limited amount of high temperature heat load in the site, which constrains the heat pump.
In the third interval we have a limited amount of low temperature heat generation in the site, which again constrains how much the heat pump can run.
With a Valve
Lets now optimize a heat pump with a high temperature to low temperature valve.
This allows heat to flow from high to low temperature, which means our boiler can generate high temperature heat that ends up as low temperature heat input into the heat pump.
This is a pretty bizarre situation, that is optimal because of our negative electricity price.
import energypylinear as epl
asset = epl.HeatPump(
electric_power_mw=1.0,
cop=2,
gas_prices=20,
electricity_prices=[-100, -100, -100],
high_temperature_load_mwh=[3.0, 0.5, 3.0],
low_temperature_generation_mwh=[4.0, 4.0, 0.0],
include_valve=True
)
simulation = asset.optimize(
verbose=4,
)
print(simulation.results[
[
"site-high_temperature_load_mwh",
"site-low_temperature_generation_mwh",
"spill-low_temperature_load_mwh",
"heat-pump-electric_load_mwh",
"heat-pump-high_temperature_generation_mwh",
"boiler-high_temperature_generation_mwh"
]
])
site-high_temperature_load_mwh site-low_temperature_generation_mwh spill-low_temperature_load_mwh heat-pump-electric_load_mwh heat-pump-high_temperature_generation_mwh boiler-high_temperature_generation_mwh
0 3.0 4.0 3.0 1.0 2.0 1.0
1 0.5 4.0 4.5 1.0 2.0 0.0
2 3.0 0.0 0.0 1.0 2.0 2.0
We now see that our heat pump operates for all three intervals. It is no longer starved for low temperature heat input, as the boiler can generate this heat.