Process data for Weekly Margin simulation.
You can install the package from Github like this :
# Install package from Github
remotes::install_github("rte-antares-rpackage/antaresWeeklyMargin")
library( "antaresWeeklyMargin" )
You can read Hydro XML files :
# Lecture données vallées
vallees <- read_hydro_vallee(path = "inputs/hydro/")
str(vallees)
vallees
# Lecture données usines
usines <- read_hydro_usine(path = "inputs/hydro/")
str(usines)
usines
You can read Thermal XML files :
# Lecture données tranche (puissances thermiques)
tranche <- read_thermal_tranche(path = "inputs/thermal/")
str(tranche)
tranche
You can retrieve data from Eco2mix via the Opendatasoft API, for now only national real time data can be retrieved :
eco2mix <- get_eco2mix(
from = "2018-01-06",
to = "2018-01-12"
)
Specificly you can access hydraulique au fil de l'eau éclusée data like this :
# Get previous friday
vendredi <- get_previous("vendredi")
# Get previous saturday before friday
samedi <- get_previous("samedi", vendredi)
# Get data
fil_eau <- get_hydraulique_fil_de_l_eau_eclusee(from = samedi, to = vendredi)
date | date_heure | hydraulique_fil_de_l_eau_eclusee |
---|---|---|
13/01/2018 | 13/01/2018 01:00 | 5954 |
13/01/2018 | 13/01/2018 02:00 | 5909 |
13/01/2018 | 13/01/2018 03:00 | 5702 |
13/01/2018 | 13/01/2018 04:00 | 5633 |
13/01/2018 | 13/01/2018 05:00 | 5637 |
Access to an API can be blocked by the proxy, you can set your credentials to allow retrieveng data :
setupProxy(user = "NNI", proxy_pwd = "MOT_DE_PASSE")
# You need to restart your R session
get_proxy_info() # check if settings are correctly set
It only needs to be done once. Don't forget to restart your R session for change to be effective !
You can retrieve NTC data from RTE data API with function get_ntc
.
This API require to have an account and to subscribe to a specific API (Abonnez-vous à l'API) and creating an application.
Before retrieving data, you need a token :
id_client <- "XXXXX-XXXXX-XXXXX-XXXXX-XXXXX"
id_secret <- "XXXXX-XXXXX-XXXXX-XXXXX-XXXXX"
token <- get_token(
key = list(id_client = id_client, id_secret = id_secret)
)
id_client
and id_secret
can be found on the page Mes applications if you're logged.
With this token you can call get_ntc
:
ntc <- get_ntc(token = token, type = c("ANNUAL", "MONTHLY"))
type | sender_country_name | sender_country_eic_code | receiver_country_name | receiver_country_eic_code | start_date | end_date | updated_date | value |
---|---|---|---|---|---|---|---|---|
MONTHLY | Germany | 10YCB-GERMANY--8 | France | 10YFR-RTE------C | 01/02/2018 | 02/02/2018 | 18/01/2018 | 1200 |
MONTHLY | Germany | 10YCB-GERMANY--8 | France | 10YFR-RTE------C | 02/02/2018 | 03/02/2018 | 18/01/2018 | 1200 |
MONTHLY | Germany | 10YCB-GERMANY--8 | France | 10YFR-RTE------C | 03/02/2018 | 04/02/2018 | 18/01/2018 | 1200 |
MONTHLY | Germany | 10YCB-GERMANY--8 | France | 10YFR-RTE------C | 04/02/2018 | 05/02/2018 | 18/01/2018 | 1200 |
MONTHLY | Germany | 10YCB-GERMANY--8 | France | 10YFR-RTE------C | 05/02/2018 | 06/02/2018 | 18/01/2018 | 1200 |
MONTHLY | Germany | 10YCB-GERMANY--8 | France | 10YFR-RTE------C | 06/02/2018 | 07/02/2018 | 18/01/2018 | 1200 |