The ‘antaresRead’ Package

François Guillem

2018-02-23

This document describes a typical use of the antaresRead package.

This package has been designed to read the results of an Antares simulation in an easy and convivial way. It can of course read any output files of a simulation at any desired time step, read synthetic results or detailed Monte Carlo scenarios but it can also add input time series to the results and perform some basic treatments like removing virtual areas.

Installation

The antaresRead package depends on the packages data.table, plyr and lubridate. If you have not already got them you can install them with the following command:

install.packages(c("data.table", "plyr", "lubridate"))

Then you can install the antaresRead package either with the Rstudio assistant in the “Packages” tab or with the following command:

install.packages("path_to_the_package/antaresRead_***.zip")

Before reading data

First, when you start a new R session, you need to load the package:

library(antaresRead)

To display the list of all the functions of the package and access their help pages, type in the R console:

help(package = "antaresRead")

Then you can start using the package. The first function to use is setSimulationPath.

This function needs to be called at least once during each R session. It stores important informations that are used by most of the functions of the package. While it has not been run, these functions will not work.

Without any argument, setSimulationPath asks interactively to choose a directory containing an antares study. If the study contains multiple simulation results, it will also asks the user to choose one of them. This function stores the path to the output and reads some useful information about the simulation: type of output available, list of areas, links and clusters in the simulation, variables present in the output files, etc.

setSimulationPath can also be used in a non-interactive way with one of these syntaxes:

# Specify full path
setSimulationPath("study_path/output/simulation_name")

# Specify the name of the simulation
setSimulationPath("study_path", simulation_name)

# Select a simulation by order
setSimulationPath("study_path", 1) # first simulation

# Select a simulation by reverse order
setSimulationPath("study_path", -1) # last simulation

# It is possible to store in a variable the result of the function
opts <- setSimulationPath("study_path", 1)

The function returns an object containing informations about the selected simulation. You can store this object in a variable for later use but this is not necessary because at any moment you can retrieve these informations.

Reading time series

Once setSimulationPath has been run, you can start reading data. Function readAntares is there for that !

readAntares is the main function of the package. It is used to read every possible time series and it performs a few treatments on them to make your life easier. The result of the function will have the simplest structure possible: either a simple table or a list of tables if you asks data for differents elements (for instance links and areas)

It has a huge number of parameters to control exactly what you get, but all of them are optional. Without any argument the function will still works and it will read the synthetic results for all the areas. But you can import other kind of output. here are some examples:

# Synthetic results for all links
readAntares(links="all")

#Synthetic results for all clusters
readAntares(clusters="all")

# Areas and links at the same time
readAntares(areas="all", links="all")

# Select only a few columns. 
readAntares(select = c("OV. COST", "OP. COST", "LOAD"))

You can also choose what elements to import and what level of details you want. For instance, the following command reads the first 10 Monte-Carlo scenarii data at monthly time step for the areas named “area1”, “area2” and “area3”.

readAntares(areas=c("area1", "area2", "area3"), timeStep="monthly",
            synthesis=FALSE, mcYears = 1:10)

Finally many arguments of readAntares can be used to add input time series to the object returned by the function. For instance, misc=TRUE will add columns containing miscelaneous productions for the imported areas.

Manipulating data

readAntares returns either a single table or a list of tables depending on the query of the user. More precisely the tables are data.table objects. It is then possible to use the powerful syntax offered by the package data.table.

The general syntax is like:

name_of_the_table[filter_rows, select_columns, group_by]

For instance, areas[area == "08_fr", .(timeId, LOAD)] will return a table containing columns timeId and LOAD for the area names “08_fr”. In the select statement, it is also possible to calculate new columns. For instance, one can compute the net load like this:

areas[, .(area, timeId, netLoad = LOAD - `ROW BAL.` - PSP - MISC. - NDG -
                                    H. ROR - WIND - SOLAR)]

One can also compute agregated statistics. For instance, the following code will compute the total load of all areas per timeId:

areas[, .(totalLoad = sum(LOAD)), by = .(timeId)]

Of course, agregation also works with filters. For instance to compute the total load only for french areas (assuming their names contain “fr”):

areas[area %in% getAreas("fr"), .(totalLoad = sum(LOAD)), by = .(timeId)]

If you are not familiar with package data.table, you should have a look at the documentation and especially at the vignettes of the package:

help(package="data.table")

Other useful functions

readAntares can import almost everything but not everything because some data is not time series. Other functions exist to read this specific data: readBindingConstraints to read binding constraints, readClusterDesc to read cluster characteristics and readLayout to get the coordinates of the areas in the user interface of Antares.

Some parameters in readAntares and other functions wait for vectors of area names or link names. On large projects with lots of areas. It may be painful to specify by hand a long list of areas or links. Hopefully, the functions getAreasand getLinks can be used to select or exclude areas using regular expressions. For instance, let us assume that the name of all areas located in France start with the characters“fr”, then the following command returns the list of all french areas:

getAreas("fr")

To exclude offshore production areas (assuming their name contains the word “offshore”) one can use:

getAreas("fr", exclude="offshore")

A few other functions are provided by the package. To see a list of them, type in the console:

help(package = "antaresRead")