Benchmark test for heat sector

[ursulaelmir]

Fix #503

Changes proposed in this pull request:

• Prepare input files for electric grid + heat pump + heat grid
• Write benchmark test
• Write definitions for all benchmark tests

The following steps were realized, as well (if applies):

• Use in-line comments to explain your code
• Write docstrings to your code
• For new functionalities: Explain in readthedocs
• Write test(s) for your new patch of code
• Update the CHANGELOG.md
• Apply black (black . --exclude docs/)
• Check if benchmark tests pass locally (EXECUTE_TESTS_ON=master pytest)

Please mark above checkboxes as following:

• Open
• Done

❌ Check not applicable to this PR

_{For more information on how to contribute check the CONTRIBUTING.md.}

[ursulaelmir]

Hi @smartie2076 , I was thinking of integrating a time series for the electricity DSO in this benchmark test, so it would also look into #506

Why I am thinking about that, it's basically because in this simple test with electricity DSO, heat DSO and heat pump, the heat pump will only be used when electricity_price/COP < heat_price
so I can have a time series of only two values and can check that the heat pump is only used when this is valid.
What do you think?

I am actually trying to implement it myself but I get this error: json.decoder.JSONDecodeError: Expecting property name enclosed in double quotes: line 1 column 2 (char 1)
when energy_price for electricity DSO is "{0.1 , 0.3}"

[ursulaelmir]

or this could also be implemented as a sensitivity analysis @smartie2076 @Bachibouzouk

[ursulaelmir]

@smartie2076 I have written the benchmark tests which are ready for review. Please let me know if we will include the time series here or not :)

[smartie2076]

Hi Ursula! Actually, you should be able to implement a timeseriers with "{‘file_name’: ‘your_file_name.csv’, ‘header’: ‘your_header’, ‘unit’: ‘your_unit’}". You can not use two values in the timeseries, the series has to be as long as the others.

The idea is good!

(readthedocs says "{‘value’: {‘file_name’: ‘your_file_name.csv’, ‘header’: ‘your_header’}, ‘unit’: ‘your_unit’}", but I don't think that's true)

[smartie2076]

Including the quotation marks!

[ursulaelmir]

hi @smartie2076 , I tried both ways and the error is still the same: json.decoder.JSONDecodeError: Expecting property name enclosed in double quotes: line 1 column 2 (char 1) .. I am not sure what the problem is..

[smartie2076]

Okay, please push with the input files and I look into it tomorrow.

[smartie2076]

The dict has to be defined using string delimiter ', otherwise the data can not be parsed.

A new error occurs now with the function C3.check_feedin_tariff() as that function can not process timeseries yet. You can write something like

If value in feedin-tariff or value in electricity price:
Diff = fit - ep
If any(diff) > 0:
Warning 

It might be that this only works for mixed problems (one series, one scalar) if the scalar is of type int. Please check if it can be float (because it will most likely be a float). Maybe use pd.multiply or something then.

[ursulaelmir]

hi @smartie2076
I am not so sure if what I am doing is what you're looking for.. something like that? (of course it is not done.. but errors accumulate)

    for provider in dict_values[ENERGY_PROVIDERS].keys():
        feedin_tariff = dict_values[ENERGY_PROVIDERS][provider][FEEDIN_TARIFF]
        electricity_price = dict_values[ENERGY_PROVIDERS][provider][ENERGY_PRICE]
        if "value" in feedin_tariff and "value" in electricity_price:
            diff = feedin_tariff["value"] - electricity_price["value"]
            if diff > 0:
                msg = f"Feed-in tariff > energy price of energy provider asset '{dict_values[ENERGY_PROVIDERS][provider][LABEL]}' would cause an unbound solution and terminate the optimization."
                raise ValueError(msg)
        elif "value" in feedin_tariff and "flow" in electricity_price:
            for i in range(0, electricity_price["flow"]["data"]):
                diff[i] = feedin_tariff["value"] - electricity_price["flow"]["data"][i]
                if any(diff) > 0:
                    msg = f"Feed-in tariff > energy price of energy provider asset '{dict_values[ENERGY_PROVIDERS][provider][LABEL]}' would cause an unbound solution and terminate the optimization."
                    raise ValueError(msg)

[Bachibouzouk]

hi @smartie2076
I am not so sure if what I am doing is what you're looking for.. something like that? (of course it is not done.. but errors accumulate)

@ursulaelmir if you set the code between ```python

and ```
it will highlight it like a python editor ;)

[smartie2076]

Yeah, something like that! Please always use global variables:

    for provider in dict_values[ENERGY_PROVIDERS].keys():
        feedin_tariff = dict_values[ENERGY_PROVIDERS][provider][FEEDIN_TARIFF]
        electricity_price = dict_values[ENERGY_PROVIDERS][provider][ENERGY_PRICE]
        if VALUE in feedin_tariff and VALUE in electricity_price:
            diff = feedin_tariff[VALUE] - electricity_price[VALUE]
            if diff > 0:
                msg = f"Feed-in tariff > energy price for the energy provider asset '{dict_values[ENERGY_PROVIDERS][provider][LABEL]}' would cause an unbound solution and terminate the optimization. Please reconsider your feed-in tariff and energy price."
                raise ValueError(msg)
        else: 
            diff = feedin_tariff - electricity_price
            boolean = [k > 0 for k in diff.values] # True if there is an instance where feedin tarrif > electricity_price
            if any(boolean) == True:
               instances = sum(boolean) # Count instances
                msg = f"Feed-in tariff > energy price in {instances} during the simulation time for the energy provider asset '{dict_values[ENERGY_PROVIDERS][provider][LABEL]}'. This would cause an unbound solution and terminate the optimization. Please reconsider your feed-in tariff and energy price."
                raise ValueError(msg)

[ursulaelmir]

at the moment there's a keyError 0 and when I print electricity price, it's
{'file_name': 'tariff_timeseries.csv', 'header': 'currency/kWh', 'unit': 'currency/kWh'}
for the electricity dso. not sure how to access the csv values.. @smartie2076

[smartie2076]

Okay, nice that we have the feature of energy price tariffs tested now as well, combined with the electricity sector.

I would propose though that we split the benchmark tests into multiple files: By now I wonder what "simple_scenarios" are. I think a sector coupling example is not necessarily a simple scenario.

Also, we should add docstrings to our benchmark tests, so that we know what they are about and how we might have to fix them.

[Bachibouzouk]

By now I wonder what "simple_scenarios" are. I think a sector coupling example is not necessarily a simple scenario.

I view the name "simple scenario" as "the simplest possible case of sector coupling". Even though sector coupling might be considered more complicated than just PV + battery.

[ursulaelmir]

@smartie2076 I added some definitions for the benchmark tests and removed the word "simple" for now. Is this okay?

[smartie2076]

By now I wonder what "simple_scenarios" are. I think a sector coupling example is not necessarily a simple scenario.

I view the name "simple scenario" as "the simplest possible case of sector coupling". Even though sector coupling might be considered more complicated than just PV + battery.

Hm, but the file includes all benchmark tests for the electricity sector (#106), as well as for the heat sector - basically sector-coupled cases - (#503), and also peak demand pricing...

I probably understand it better if we have following benchmark test groups in seperate files:

• energy system modelling ("simple scenarios")
• extreme scenarios and validation
• features (peak demand pricing, timeseries instead of scalars for efficiencies)
• KPI (investment optimization, expected KPI)
• Constraints (minimal renewable share, net zero, co2)

...@Bachibouzouk are these too many files? I am a bit peculiar in organising my files, so I am not sure.

[Bachibouzouk]

...@Bachibouzouk are these too many files? I am a bit peculiar in organising my files, so I am not sure.

I don't think it is too many files, it is fine :)

[smartie2076]

Two suggestions, otherwise ready to merge!

We can rename the benchmark tests according to my suggestions in another PR. :)

[ursulaelmir]

@smartie2076 there is an error I think related to merging dev into this branch. am I missing something?

[smartie2076]

@smartie2076 there is an error I think related to merging dev into this branch. am I missing something?

Have you added the required file constraints.csv to your benchmark tests? Is the key minimal_renewable_share in there? I got that hint from this error message:

E                   AssertionError: Key minimal_renewable_share is not in extra_parameters, but is in missing_parameters in the folder /home/travis/build/rl-institut/mvs_eland/tests/benchmark_test_inputs/AFG_grid_heatpump_heat.

E                   assert 'minimal_renewable_share' in {'maximumCap': {'default': None, 'required in files': ['energyConversion', 'energyProduction'], 'unit': 'None', 'warni...s': ['energyProviders'], 'unit': 'str', 'warning_text': 'defines the unit of power provided by a DSO (Values: str). '}}

[ursulaelmir]

@smartie2076 there is an error I think related to merging dev into this branch. am I missing something?

Have you added the required file constraints.csv to your benchmark tests? Is the key minimal_renewable_share in there? I got that hint from this error message:

E                   AssertionError: Key minimal_renewable_share is not in extra_parameters, but is in missing_parameters in the folder /home/travis/build/rl-institut/mvs_eland/tests/benchmark_test_inputs/AFG_grid_heatpump_heat.

E                   assert 'minimal_renewable_share' in {'maximumCap': {'default': None, 'required in files': ['energyConversion', 'energyProduction'], 'unit': 'None', 'warni...s': ['energyProviders'], 'unit': 'str', 'warning_text': 'defines the unit of power provided by a DSO (Values: str). '}}

THANKS!