Add VDI profiles based on lpagg #50
Conversation
Features/revise methods for elec slp
…s/add-vdi-from-lpagg
…mof/demandlib into features/add-vdi-from-lpagg
|
Implementation of a leap year is still missing but everything else should be fine now. We may want to merge this and start a new PR just for implementing leap years. @jnettels In my opinion it is not a good idea to create a second PR. I merged both but it cost me some time. |
There was a problem hiding this comment.
Thanks for your effort!
I wanted to give this feature another chance of merging, but I had to spend my time adding missing code, tests and docs in an efficient way. I needed a clean base to do that. Also after >2.5 years I could not assume there would still be interest in this specific PR. This PR is targeting v0.2dev, which has long been surpassed*, and the branch was out of sync with dev in many regards. So the most efficient way for me seemed to start a fresh branch and cherry-pick all the necessary commits, which worked fine. It produced a clean history, passed all tests and could have just been merged. For me it was the choice between doing it like this, or not contributing at all (because I could not handle all the changes unrelated to the actual new feature). But I do not mind removing my PR, as long as we get the feature merged.
Now for some feedback:
- You removed the
file_weatherargument ofRegion()and I do not see an obvious way to use custom weather data. We could add a new functionfrom_file()toClimate(), whichfrom_try_data()could also use internally. - The docs need to be updated with usage of the
Climate()class - My code supported houses where
Q_Heiz_a,W_aorQ_TWW_aare not defined. I think this is important, because sometimes you may e.g. only want the heating profiles, but no electricity profile. With my proposed changes, those columns are returned as all NaN, which can easily be filtered by the user. demandlib.iniandconfig.pyare still present. You used them at one point to store the default values forsummer_temperature_limitandwinter_temperature_limit. Now it only containstest_entry = 5, which I assume is also unrelated to the VDI feature. Are these files actually used?- Footnote*: I am very confused. Is
devthe actual dev branch, orv0.2dev? The latter has updates to the BDEW api (get_scaled_power_profiles()instead ofget_profile()), which are now also mixed in with this PR, making it even more confusing...
| q_heiz_a = house["Q_Heiz_a"] | ||
| w_a = house["W_a"] | ||
| q_tww_a = house["Q_TWW_a"] |
There was a problem hiding this comment.
Allow skipping any energy type
| q_heiz_a = house["Q_Heiz_a"] | |
| w_a = house["W_a"] | |
| q_tww_a = house["Q_TWW_a"] | |
| q_heiz_a = house.get("Q_Heiz_a", float("NaN")) | |
| w_a = house.get("W_a", float("NaN")) | |
| q_tww_a = house.get("Q_TWW_a", float("NaN")) |
| # The typical day calculation inherently does not add up to the | ||
| # desired total energy demand of the full year. Here we fix that: | ||
| for column in load_curve_house.columns: | ||
| q_a = house[column.replace("TT", "a")] |
There was a problem hiding this comment.
Allow skipping any energy type
| q_a = house[column.replace("TT", "a")] | |
| q_a = house.get(column.replace("TT", "a"), float("NaN")) |
| def from_try_data(self, try_region, hoy=8760): | ||
| if try_region not in list(range(1, 16)): | ||
| raise ValueError( | ||
| f">{try_region}< is not a valid number of a DWD TRY region." | ||
| ) | ||
| fn_weather = os.path.join( | ||
| os.path.dirname(__file__), | ||
| "resources_weather", | ||
| "TRY2010_{:02d}_Jahr.dat".format(try_region), | ||
| ) | ||
| weather = dwd_try.read_dwd_weather_file(fn_weather) | ||
| weather = ( | ||
| weather.set_index( | ||
| pd.date_range( | ||
| datetime.datetime(2010, 1, 1, 0), | ||
| periods=hoy, | ||
| freq="h", | ||
| ) | ||
| ) | ||
| .resample("D") | ||
| .mean() | ||
| ) | ||
| self.temperature = weather["TAMB"] | ||
|
|
||
| weather.loc[weather["CCOVER"] >= 5, "cloud_category"] = "B" | ||
| weather.loc[weather["CCOVER"] < 5, "cloud_category"] = "H" | ||
| self.cloud_coverage = weather["cloud_category"] | ||
| fn_energy_factors = os.path.join( | ||
| os.path.dirname(__file__), | ||
| "vdi_data", | ||
| "VDI_4655_Typtag-Faktoren.csv", | ||
| ) | ||
| self.energy_factors = pd.read_csv( | ||
| fn_energy_factors, | ||
| index_col=[0, 1, 2], | ||
| ).loc[try_region] | ||
| return self |
There was a problem hiding this comment.
Allow custom DWD weather data by exposing from_file().
| def from_try_data(self, try_region, hoy=8760): | |
| if try_region not in list(range(1, 16)): | |
| raise ValueError( | |
| f">{try_region}< is not a valid number of a DWD TRY region." | |
| ) | |
| fn_weather = os.path.join( | |
| os.path.dirname(__file__), | |
| "resources_weather", | |
| "TRY2010_{:02d}_Jahr.dat".format(try_region), | |
| ) | |
| weather = dwd_try.read_dwd_weather_file(fn_weather) | |
| weather = ( | |
| weather.set_index( | |
| pd.date_range( | |
| datetime.datetime(2010, 1, 1, 0), | |
| periods=hoy, | |
| freq="h", | |
| ) | |
| ) | |
| .resample("D") | |
| .mean() | |
| ) | |
| self.temperature = weather["TAMB"] | |
| weather.loc[weather["CCOVER"] >= 5, "cloud_category"] = "B" | |
| weather.loc[weather["CCOVER"] < 5, "cloud_category"] = "H" | |
| self.cloud_coverage = weather["cloud_category"] | |
| fn_energy_factors = os.path.join( | |
| os.path.dirname(__file__), | |
| "vdi_data", | |
| "VDI_4655_Typtag-Faktoren.csv", | |
| ) | |
| self.energy_factors = pd.read_csv( | |
| fn_energy_factors, | |
| index_col=[0, 1, 2], | |
| ).loc[try_region] | |
| return self | |
| def from_try_data(self, try_region, hoy=8760): | |
| if try_region not in list(range(1, 16)): | |
| raise ValueError( | |
| f">{try_region}< is not a valid number of a DWD TRY region." | |
| ) | |
| fn_weather = os.path.join( | |
| os.path.dirname(__file__), | |
| "resources_weather", | |
| "TRY2010_{:02d}_Jahr.dat".format(try_region), | |
| ) | |
| self.from_file(fn_weather, try_region, hoy) | |
| return self | |
| def from_file(self, fn_weather, try_region, hoy=8760): | |
| if try_region not in list(range(1, 16)): | |
| raise ValueError( | |
| f">{try_region}< is not a valid number of a DWD TRY region." | |
| ) | |
| weather = dwd_try.read_dwd_weather_file(fn_weather) | |
| weather = ( | |
| weather.set_index( | |
| pd.date_range( | |
| datetime.datetime(2010, 1, 1, 0), | |
| periods=hoy, | |
| freq="h", | |
| ) | |
| ) | |
| .resample("D") | |
| .mean() | |
| ) | |
| self.temperature = weather["TAMB"] | |
| weather.loc[weather["CCOVER"] >= 5, "cloud_category"] = "B" | |
| weather.loc[weather["CCOVER"] < 5, "cloud_category"] = "H" | |
| self.cloud_coverage = weather["cloud_category"] | |
| fn_energy_factors = os.path.join( | |
| os.path.dirname(__file__), | |
| "vdi_data", | |
| "VDI_4655_Typtag-Faktoren.csv", | |
| ) | |
| self.energy_factors = pd.read_csv( | |
| fn_energy_factors, | |
| index_col=[0, 1, 2], | |
| ).loc[try_region] | |
| return self |
|
Okay, I agree the PR is allready very complex so I see two things we can do:
I would strongly prefere the second variant. Nothing happend by chance and I have good reasons why I did what I did. We will merge the PR to |
|
Because of the difficult merge conflicts it looks more as it is. But this is one more reason to merge it and start with a cleaner structure. Thank you for all you comments and contributions. |
The basic code is taken from https://github.com/jnettels/lpagg but most of the code has been revised for modular use and to speed up the code. The lpagg package contains various functionality but only the VDI-profiles and reading the TRY data from DWD is used.
To test the new profiles use the
vdi_profile_examplein the examples section. The parameter of the houses are added within a loop. I used this to multiply the number of houses to check how long it would take to model e.g. 1000 houses. 1000 houses on an hourly base will take less than a minute, which shows that the draft is at least fast enough 🏎️Related to #36
Install this branch to test the code:
It is the first draft, I am open for concrete ideas.