Ctrl+K

History of Changes

History of Changes#

Version 0.5.12 (07.04.2024)#

  • fix the filtering of GCPT/GEM coal database by Status.

  • add technology renaming for GWPT/GEM wind dataset.

  • improve performances of GEM data processing.

Version 0.5.11 (05.02.2024)#

  • The default configuration no longer filters out retired power plants. This filtering should be done by the user afterwards.

Version 0.5.10 (30.01.2024)#

  • fix deprecation warnings for new pandas version for inplace operations

  • fix bus mapping in export module

Version 0.5.9 (16.01.2024)#

  • fix deprecation warnings for new pandas version

  • lower GEM reliability score to fix DateIn and DateOut for some powerplants

Version 0.5.8 (30.10.2023)#

New Features

  • Datasets from the Global Energy Monitor are now combined into one dataset called GEM. This is per default used in the matching process.

  • Updates the following Global Energy Monitor data according to latest May 2023 release: * GSPT, solar power plant * GWPT, wind power plant

  • Changing Global Energy Monitor dataset name to -latest to avoid data update PRs in powerplantmatching but rather encourage updates in Global Energy Monitor data repo.

Fixes

  • In the aggregation and merge process, the DateIn and DateOut columns are now aggregated by taking the minimum and maximum value respectively.

Version 0.5.7 (30.05.2023)#

Bug fixes

  • Replace unstable FIAS and OPSD download links by new provider

  • Replace ValueError when no matches found to logged warning.

Other Changes

  • adds the following new Global Energy Monitor data including: * GBPT, bioenergy power plant * GGPT, gas power plant update * GCPT, coal power plant * GGTPT, geothermal power plant * GNPT, nuclear power plant * GSPT, solar power plant * GWPT, wind power plant * GHPT, hydro power plant

Version 0.5.6 (13.02.2023)#

Bug fixes

  • Set DateIn column as a float

  • Remove special characters from hash path for custom configs

  • Add conversion of country name into GEO and GPD data sources

Other Changes

  • Revise table in documentation and update links to powerplant database

Version 0.5.5 (21.12.2022)#

New Features

  • New EXTERNAL_DATABASE interface to integrate additional custom data of raw data matching the powerplantmatching format.

  • The example notebook was updated to the new version of the package.

  • Rename the matching_analysis directory to analysis.

  • Harmonize the analysis scripts with the pm.powerlants(update=True) functionality.

Bug fixes

  • Fix GGPT (formerly GEM_GGPT) interface.

  • Fix missing column names in dataframes

  • Spanish hydro stores with an capacity larger than 50GWh in the JRC data base are assumed to be reservoirs even if stated differently.

  • add geolocations for Italian geothermal units and adjust filters in config.yaml to bring geothermal to powerplants.csv

Other Changes

  • Rename the matching_analysis directory to analysis.

  • Rename GEM_GGPT to GGPT and add GEM_GGPT as an deprecated alias.

  • Rename matched_data to powerplants in collection.py and add matched_data as an deprecated alias.

  • Fueltype “Other” was remove from the Capacity_Stats function.

Version 0.5.4 (02.08.2022)#

New Features

  • The GEM_GGPT data containing global gas power plant data for all countries is now available.

Bug fixes

  • Fix capacity-weighted mean calculation of efficiencies. The previous implementation lead to underestimated efficiencies for power plants consisting of multiple power units.

Version 0.5.3 (08.04.2022)#

  • Bug fix for renewable power plants from OPSD

Version 0.5.2 (07.04.2022)#

  • The overall config setting was fine-tuned in order to improve the matching results.

  • New scripts were added to the folder matching_analysis

Version 0.5.1 (04.04.2022)#

New Features

  • The IRENASTATS data is now available containing powerplant capacities for all countries from 2000 to 2020.

Bug fix

  • The url of the powerplants function was fixed.

Other

  • The removal of the column DateMothBall was caught up on.

  • The manual corrections were reactivated.

  • Improved country code and name conversion by using country_converter.

Version 0.5 (04.04.2022)#

This release contains many breaking changes. Due to time-constraints we cannot ensure a smooth transition to the new release. If you are using a custom config file (e.g. ~/powerplantmatching_config.yaml) please be aware of the following config changes:

Configuration Changes

  • The custom configuration now only updates the package default configuration, which makes the compatibility of configuration updates much easier. So, instead of replacing the whole package configuration (the default config provided by powerplantmatching), the new purpose of the custom config is to adjust individual values. So, please make sure to only add keys to the custom config which you want to change in comparison to the default config.

  • The following sections of the configuration file ~/powerplantmatching_config.yaml changed: * the target_fueltypes section is now mapping the representative fueltypes to the regular expressions that are used in order to determine them. * the target_technologies section is now mapping the representative technologies to the regular expressions that are used in order to determine them. * the target_set section is now mapping the representative sets to the regular expressions that are used in order to determine them. * a section clean_name was added. This section contains the regular expressions and lists of words that are used to clean the names of the plants.

In order to ensure compatibility with the new code, please delete these sections in your custom config.

Deprecations

  • The CARMA dataset was deprecated as the data is not publicly available anymore.

  • The IWPDCY dataset was deprecated in the favor of the JRC data.

  • The WEPP dataset was deprecated due to restrictive license.

  • The UBA dataset was deprecated in the favor of the OPSD data.

  • The BENTZA dataset was deprecated in the favor of the OPSD data.

  • The IRENA_stats dataset was deprecated as the data is not publicly available anymore.

  • The following functions were deprecated and will be removed in v0.6: * powerplantmatching.export.to_TIMES * powerplantmatching.export.store_open_dataset * powerplantmatching.export.fueltype_to_abbrev` ` * ``powerplantmatching.heuristics.set_denmark_region_id * powerplantmatching.heuristics.remove_oversea_areas * powerplantmatching.heuristics.set_known_retire_years

  • The argument extendby_kwargs in the function powerplantmatching.collection.matched_data was deprecated in the favor of extend_by_kwargs.

New Features

  • The BEYOND COAL data is now available as an data source.

  • A new dataset WIKIPEDIA on nuclear powerplants in europe from wikipedia was added.

  • The GEO dataset returns powerplant blocks instead of whole plants.

  • All scripts were aligned with the black coding style.

  • A documentation on readthedocs was added.

  • The config has now a key main_query which is applied to all datasets.

  • A CI was added.

  • A new function powerplantmatching.heuristics.isin was added. It checks which data entries of a non-matched dataset is included in a matched dataset.

Breaking Code Changes:

  • The argument rawDE and rawEU in powerplantmatching.data.OPSD was deprecated in the favor of raw. If True the function returns a dictionary with the raw datasets.

  • All keyword arguments of the data functions in powerplantmatching.data were sorted according to raw, update, config. This lead to some breaking changes of the arguments order.

  • The Fueltype Other was replaced by NaN.

  • The GEO data now returns a dataset containing power plant units.

  • The ESE dataset was removed due the hosting website taken down.

  • The argument subsume_uncommon_fueltypes_to_other in powerplantmatching.collection.matched_data was removed.

  • The function powerplantmatching.cleaning.aggregate_units does not support the arguments use_saved_aggregation and save_aggregation anymore due to it’s insecure behavior.

  • The function powerplantmatching.matching.compare_two_datasets does not support the arguments use_saved_matches anymore due to it’s insecure behavior.

Version 0.4.6 (25.11.2020)#

Triggered by the ongoing phase-outs of especially nuclear, coal and lignite plants in many countries, we acknowledge that time-related data of power stations and their single blocks becomes increasingly important.
Therefore, we decided to - adapt the columns: - rename YearCommissioned to DateIn (reflects when a station/block had initially started operation) - rename Retrofit to DateRetrofit (reflects when a station/block has been retrofitted) - add DateMothball (reflects when a station/block has been mothballed) - add DateOut (reflects when a station/block has been finally decommissioned)
Please note: Currently, these columns only contain the year, but we aim in future to provide exact dates (i.e. including day and month) wherever possible.
Further changes: - new target_columns: - add EIC (the European Energy Identification Code) - add StorageCapacity_MWh - update JRC Hydro Database to v5 and add quick workaround so that pm can deal with non-unique identifiers - replace deprecated by current pandas functions - custom configuration and package configuration are now merged when calling get_config() (values in the custom configuration are prioritized) - fix retrieving BNETZA data
- export.py has now a function for mapping bus coordinates to the power plant list

Version 0.4.1 (02.08.2019)#

Data structure#

  • abolish git lfs in the favour of direct url parsing

  • store data in user folders

    • Linux ~/.local/share/powerplantmatching

    • Windows C:\Users\<USERNAME>\AppData\Roaming\powerplantmatching

  • move necessary files to package_data in powerplantmatching folder (such as duke binaries, xml files etc.)

  • include JRC Hydro Database

Code#

  • get rid of mutual module imports

  • speed up grouping (cleaning.py, matching.py)

  • revise/rewrite code in data.py

  • enable switch for matching powerplants of the same country only (is now default, speeds up the matching and aggregation process significantly)

  • boil down plot.py which caused long import times

  • get rid of config.py in the favour of core.py and accessor.py

  • drop deprecated functions in collection.py which now only includes collect() and matched_data()

Contents