Python client for the ENTSO-G API (european network of transmission system operators for gas)
Documentation of the API found on https://transparency.entsog.eu/api/archiveDirectories/8/api-manual/TP_REG715_Documentation_TP_API%20-%20v2.1.pdf
Documentation of the data (user manual) found on https://www.entsog.eu/sites/default/files/2021-07/ENTSOG%20-%20TP%20User%20Manual_v_4.5.pdf
Heavily inspired upon (and forked from) https://github.com/EnergieID/entsoe-py
python3 -m pip install entsog-py
The package comes with 2 clients:
EntsogRawClient: Returns data in its raw format, usually JSONEntsogPandasClient: Returns data parsed as a Pandas DataFrame
It's preferable to use the Pandas Client as this will handle most API limitations itself. However, if you want to obtain the pure raw data; you can use the raw client.
On wwww.gasparency.com you can find example use cases of the data. Almost everything there is achieved with the help of this package!
from entsog import EntsogRawClient
import pandas as pd
client = EntsogRawClient()
start = pd.Timestamp('20171201', tz='Europe/Brussels')
end = pd.Timestamp('20180101', tz='Europe/Brussels')
country_code = 'NL' # Netherlands
client.query_connection_points()
client.query_operators()
client.query_balancing_zones()
client.query_operator_point_directions(country_code)
client.query_interconnections()
client.query_aggregate_interconnections()
client.query_urgent_market_messages()
client.query_tariffs(start = start, end = end, country_code = country_code)
client.query_tariffs_sim(start = start, end = end, country_code = country_code)
operational_options = {
interruption_capacity : "Actual interruption of interruptible capacity",
allocation : "Allocation",
firm_available : "Firm Available",
firm_booked : "Firm Booked",
firm_interruption_planned : "Firm Interruption Planned - Interrupted",
firm_interruption_unplanned :"Firm Interruption Unplanned - Interrupted",
firm_technical : "Firm Technical",
gcv : "GCV",
interruptible_available : "Interruptible Available",
interruptible_booked : "Interruptible Booked",
interruptible_interruption_actual : "Interruptible Interruption Actual – Interrupted",
interruptible_interruption_planned : "Interruptible Interruption Planned - Interrupted",
interruptible_total : "Interruptible Total",
nomination : "Nomination",
physical_flow : "Physical Flow",
firm_interruption_capacity_planned : "Planned interruption of firm capacity",
renomination : "Renomination",
firm_interruption_capacity_unplanned : "Unplanned interruption of firm capacity",
wobbe_index : "Wobbe Index",
oversubscription_available : "Available through Oversubscription",
surrender_available : "Available through Surrender",
uioli_available_lt : "Available through UIOLI long-term",
uioli_available_st : "Available through UIOLI short-term"}The Pandas Client works similar to the Raw Client, with extras:
- API limitations of big requests are automatically dealt with and put into multiple calls.
- Tariffs (and simulated tariffs) can be melted into nice storable format. Instead of having row with EUR, local currency, shared currency for each seperate product, it will create a row for each.
- Operational data can be either requested as in the raw format (which requires some loading time) or in an aggregate function
query_operational_data_allwhich will aggressively request all points in Europe and a lot faster. - It's easier to navigate points, for instance if you want to check gazprom points. See below.
from entsog import EntsogPandasClient
import pandas as pd
client = EntsogPandasClient()
start = pd.Timestamp('20171228', tz='Europe/Brussels')
end = pd.Timestamp('20180101', tz='Europe/Brussels')
country_code = 'NL' # Netherlands
client.query_connection_points()
client.query_operators(country_code)
client.query_balancing_zones()
client.query_operator_point_directions()
client.query_interconnections()
client.query_aggregate_interconnections()
client.query_urgent_market_messages()
client.query_tariffs(start = start, end = end, country_code = country_code, melt = True, verbose = True)
client.query_tariffs_sim(start = start, end = end, country_code = country_code, verbose = True)
client.query_aggregated_data(start = start, end = end, country_code = country_code)
# TODO: Add interruptions...
# client.query_interruptions(start = start, end = end)
client.query_CMP_auction_premiums(start = start, end = end)
client.query_CMP_unavailable_firm_capacity(start = start, end = end)
client.query_CMP_unsuccesful_requests(start = start, end = end)
operational_options = {
'interruption_capacity' : "Actual interruption of interruptible capacity",
'allocation' : "Allocation",
'firm_available' : "Firm Available",
'firm_booked' : "Firm Booked",
'firm_interruption_planned' : "Firm Interruption Planned - Interrupted",
'firm_interruption_unplanned' :"Firm Interruption Unplanned - Interrupted",
'firm_technical' : "Firm Technical",
'gcv' : "GCV",
'interruptible_available' : "Interruptible Available",
'interruptible_booked' : "Interruptible Booked",
'interruptible_interruption_actual' : "Interruptible Interruption Actual – Interrupted",
'interruptible_interruption_planned' : "Interruptible Interruption Planned - Interrupted",
'interruptible_total' : "Interruptible Total",
'nomination' : "Nomination",
'physical_flow' : "Physical Flow",
'firm_interruption_capacity_planned' : "Planned interruption of firm capacity",
'renomination' : "Renomination",
'firm_interruption_capacity_unplanned' : "Unplanned interruption of firm capacity",
'wobbe_index' : "Wobbe Index",
'oversubscription_available' : "Available through Oversubscription",
'surrender_available' : "Available through Surrender",
'uioli_available_lt' : "Available through UIOLI long-term",
'uioli_available_st' : "Available through UIOLI short-term"
}
client.query_operational_data(start = start, end = end, country_code = country_code, indicators = ['renomination', 'physical_flow'])
# You should use this when you want to query operational data for the entirety of continental europe.
client.query_operational_data_all(start = start, end = end, indicators = ['renomination', 'physical_flow'])
# Example for if you would like to see Gazprom points.
points = client.query_operator_point_directions()
mask = points['connected_operators'].str.contains('Gazprom')
masked_points = points[mask]
print(masked_points)
keys = []
for idx, item in masked_points.iterrows():
keys.append(f"{item['operator_key']}{item['point_key']}{item['direction_key']}")
data = client.query_operational_point_data(start = start, end = end, indicators = ['physical_flow'], point_directions = keys, verbose = False)
print(data.head())