Skip to content

Latest commit

 

History

History
808 lines (573 loc) · 22 KB

trainingPresentation.md

File metadata and controls

808 lines (573 loc) · 22 KB

Training packages R / ANTARES

css: antaresTraining2017.css author: Jalal-Edine Zawam date: 29-11-2017 autosize: true

Agenda




  1. appAntaresViz
  2. R and RStudio
  3. antaresRead
  4. antaresProcessing
  5. antaresViz

appAntaresViz

title: false type: first-title-slide

appAntaresViz

Lets try appAntaresViz

type: antaresViz title: false

Lets try appAntaresViz

Unsupplied energy scenario / Data

type: text-slide

exercise/import data :

  • Run antaresVizualisation

  • Select the folder "dataTyndp"

  • Select the simulation "20170830-1049eco-reference.h5"

  • Click on "Set Simulation"

  • Imports all areas and links

  • Click on "Validate & import data"

  • Click on "Launch Analysis"

  • Click on "tsPlot"

  • Click on "H5request"

  • Import mcYear 4

Unsupplied energy scenario / tsPlot

type: text-slide

exercise :in the tab "tsPlot" and "type : monotone/times series/heat map"

  • Find the country who have the most of unsupplied energy for the mcYear 4 between "France", "Germany", "Belgium", "Netherlands", "Portugal", "Austria" and "Great Britain"
  • For this country, which month have the most of unsupplied energy
  • Find the date when the maximum of unsupplied energy is reached, in the following we will refers to this country by Mystery and the date by Ttime
  • Why there is unsupplied energy for this year in Mystery? --> We will see this thanks to prodStack

Unsupplied energy scenario / prodStack

type: text-slide

exercise : in the tab "prodStack"

  • Import mcYear 4 and 12
  • Plot the prodStack for Mystery and for the week containing Ttime for the year 4
  • Find the amount of wind production and import/export
  • Change the parameter mcYear to 12 and find the amount of wind production and import/export
  • Does Mystery import or export for the year 12 ?
  • To whom Mystery is importing or exporting for the year 12 ? --> We will see this thanks to exchangesStack

Unsupplied energy scenario / exchangesStack

type: text-slide

exercise : in the tab "exchangesStack":

  • Import mcYear 4 and 12
  • Change the "area" to Mystery
  • For the year 12, to whom Mystery is importing or exporting ?
  • For the year 4, to whom Mystery is importing or exporting ?
  • For Ttime and the year 4, are there null flows between this country and others ? if yes, why ? --> We will see this thanks to plotMap

Unsupplied energy scenario / plotMap - Layout

type: text-slide

exercise/import data : in the tab "plotMap/Current Layout":

  • import the file "dataTyndp/mapLayoutTyndp.RDS" with the button "Browse...""

Unsupplied energy scenario / Map

type: text-slide

exercise/import data : in the tab "plotMap/Map":

  • Import mcYear 4 and 12 and change the parameter mcYear to 4
  • Change the parameter DateRange for the day containing Ttime
  • In the tab "Areas", change the parameter "color"" to "LOAD"
  • Click on update
  • In the tab "Areas", change the parameter "Size" to "H. ROR", "PSP", "MISC. NDG", "WIND", "SOLAR", "NUCLEAR", "LIGNITE", "GAS", "COAL", "OIL", "MIX. FUEL", "MISC. DTG", "H. STOR"
  • Click on update

Unsupplied energy scenario / Map with miniplots

type: text-slide

exercise/import data : in the tab "plotMap/Map":

  • In the tab "Areas/miniPlot", change the parameter "areaCharType" to "pie chart", click on update
  • In the tab "Areas/miniPlot", check the box "sizeMiniPlot", click on update
  • In the tab "Areas", change the parameter "Popup" to "UNSP. ENRG", "SPIL. ENRG", "DTG MRG", "AVL DTG", "MAX MRG", "MRG. PRICE", "BALANCE", click on update
  • Click on one pie chart

Unsupplied energy scenario / Map for links

type: text-slide

exercise/import data : in the tab "plotMap/Map":

  • In the tab "Links", change the parameter "Color" to "CONG. PROP +"
  • Click on update
  • In the tab "Links", change the parameter "Width" to "FLOW LIN."
  • Click on update
  • Click on one link
  • In the tab "Links", change the parameter "Popup" to "CONG. PROP -", "MARG. COST"
  • Click on update
  • Click on one link

Unsupplied energy scenario / Questions

type: text-slide

exercise : in the tab "plotMap/Map":

  • On the map, click on play and pause when you reach Ttime
  • Is Mystery imports or exports? to whom ?
  • Are there null flows between Mystery and others ? if yes, why ?
  • Are there links not congestionned between Netherlands and others countries ? why ?
  • Change the parameter mcYear to 12
  • Click on update
  • Is Mystery imports or exports? to whom and why ?
  • It's better to compare the two maps in one screen --> We will see this thanks to Data/compare

Unsupplied energy scenario / Compare

type: text-slide

exercise/import data : in the tab "Data":

  • Add mcYear in the parameter "plotMap"
  • Click on "Launch Analysis"
  • Go to the tab "plotMap/Map"

Unsupplied energy scenario / Compare plotMap

type: toBig

exercise/import data : in the tab "plotMap/Map":

  • Import mcYear 4 and 12
  • Reconfigure the maps with the previous parameters (Areas, color, size, popup etc.) in one step
  • Click on the first graph and choose the mcYear 4
  • Click on the second graph and choose the mcYear 12
  • Click on the update button
  • Change the parameter DateRange for the day containing Ttime
  • Click on play on one map and pause when you reach Ttime
  • At Ttime , which is the most important production in Germany for year 4 and for year 12 ?
  • Same question for Great Britain
  • The future is unpredictable so we try several scenarios to have an idea about the possible futures and increase the chance to make a good decision (more thermal capacities, more renewable energy etc.) --> We will see this thanks to Data - scenarios

Unsupplied energy scenario / Data - scenarios

type: text-slide

exercise/import data : in the tab "Data":

  • Click in the tab "Import Data"
  • Select the simulation "20170112-1832eco-pascontraintenbmax.h5"
  • Click on "Set simulation"
  • Imports all areas and links
  • Click on "Validate & import data"
  • Come back to the tab "Import Data"
  • Select a new simulation "201710118-0902eco-aveccontraintenbmax.h5"
  • Click on "Set simulation"
  • Imports all areas and links
  • Click on "Validate & import data"
  • delete "mcYear" in the parameter "plotMap"
  • Click on "Launch Analysis"

Unsupplied energy scenario / Data - scenarios - tsPlot

type: text-slide

exercise : In the tab "tsPlot":

  • Import mcYear 4
  • Change DateRange to contain the all year
  • Plot the nuclear monotone for France, Germany and Great Britain
  • Are there differences between scenarios ? if yes, why ?
  • Plot the nuclear density for France for the all year
  • Are there differences between scenarios ?
  • Plot the nuclear heatmap for France
  • For which month there is the most differences between scenarios ?

antaresViz is great but why I should learn R ?

type: text-slide

. antaresViz is designed to respond to general request but not for all requests. For particular studies we need particular representations.

. You have a lot of available function in antaresPackages, it will be a lost of time (and money ?) to recode them in python, VBA, perl or any other scripting language.

. If you learn R you will be able to use scripts from ANTARES users and you will also be able to provide scripts to other. You can propose your ideas on github.

. You be able to use a lot of statical packages (10 thousand on CRAN) : kmeans, rpart, caret, FactoMineR etc.

*Today you will learn a few things form R, if you want to learn more about R and DataScience*, ***RTE can propose you others training***.

What is R?

type: text-slide

  • R is a scientific development software specialized in calculation and statistical analysis

  • R is an open source project

  • R is a cross-platform software (linux, mac, windows ...) like ANTARES

  • R is a language.

Creating objects and assigning values in R

type: code-slide

You can type an expression without assigning its value to an object

(10 + 9) * 8
[1] 152

An object can be created with the operator assign <-

firstObject <- 15 ;firstObject
[1] 15

After this assignment, the object "firstObject" contains the value 15. Another assignment to the same object will change the content.

firstObject <- 3+rnorm(n=1) ;firstObject
[1] 2.543397

Objects and memory in R

type: code-slide

List object in memory

ls()
[1] "firstObject"

A<-34;ls()
[1] "A"           "firstObject"

Remove objects from memory

ls();rm(A)
[1] "A"           "firstObject"

ls()
[1] "firstObject"

What is RStudio?

type: text-slide


RStudio is an integrated development environment (IDE) for R. It includes a console, syntax-highlighting editor that supports direct code execution, as well as tools for plotting, history, debugging and workspace management.

antaresRead

title: false type: first-title-slide

antaresRead

Install the package "antaresRead"

type: code-slide

Like any other package...

install.packages("antaresRead")

Major functions

functions
setSimulationPath Set Path to an ANTARES simulation
readAntares Read the data of an ANTARES simulation
removeVirtualAreas Remove virtual areas
showAliases Show aliases for variables
getAreas, getLinks, getDistricts Select and exclude areas, distructs and links
readClusterDesc Import clusters Description
readInputTS Read Input Time Series

get help

## ??namePackage::function
??antaresRead::readAntares

First step : set your simulation path and read the data

type: code-slide

myPath<-"E:\\ANTARES\\Exemple_antares\\2_exemple_etudes_importantes\\TYNDP\\ST2030\\ST2030"
setSimulationPath(myPath, "20170830-1049eco-reference")
Antares project 'TYNDP2018 PEMS ST2030' (E:/ANTARES/Exemple_antares/2_exemple_etudes_importantes/TYNDP/ST2030/ST2030)
Simulation 'Reference'
Mode Economy

Content:
 - synthesis: TRUE
 - year by year: TRUE
 - MC Scenarios: FALSE
 - Number of areas: 73
 - Number of districts: 0
 - Number of links: 210
 - Number of Monte-Carlo years: 34

suppressWarnings(
  myData<-readAntares(
  areas = "all", 
  links="all", 
  clusters = "all", 
  linkCapacity = TRUE,
  mustRun = TRUE,
  showProgress = FALSE
)
)

Consolidated data : Remove your virtual areas and correct the initial data

type: first-slide

What does this function ?

??antaresRead::removeVirtualAreas
myConData<-removeVirtualAreas(
   x=myData,
   storageFlexibility = getAreas(select = c("pum", "tur")),
   production = getAreas(select = c("z_dsr", "y_mul"))
 )

getAreas and getLinks

type: title-slide

exercise :

  • Compute the number of areas without virtual areas

  • Compute the number of links without virtual areas

get help:

??antaresRead::getAreas
??antaresRead::getLinks

Data.table

type: title-slide

  • DT[ i , j , by ]

  • subset rows using i

  • then calculate j

  • grouped by

You can learn the basics about data.table with the vignette.

You can also found the cheat sheet here.

antaresDataList : List of date.tables

type: title-slide

  • readAntares return a list of data.tables.

  • Data.table is a list of vector with equal length.

exercise :

  • Compute the number of areas with unsupplied energy

  • Compute the number of areas with spilled energy

  • Compute the number of links with congestion

antaresDataList : use data.table

type: title-slide

exercise :

  • Compute the sum of the spilled energy by area and order the result by the spilled energy

  • Compute the sum of the unsupplied energy by area and order the result by the unsupplied energy

Explain the result for unsupplied energy in France: with rpart or others statical packages FactoMineR, kmeans

type: explainTheResult

Explain the result for unsupplied energy in France without some variables

type: explainTheResult2

Explain the result with some plots

title:no type:first-title-slide

Explain the result with some plots

antaresViz : prodStack, exchangesStack and plot

type: title-slide

exercise :

  • Get the first year and the first date where there is more than 700MW of unsupplied energy in France and use prodStack to visualize the production stack for this week.

  • Visualise the evolution of the echanges between France and other countries with exchangesStack.

  • Plot the price for this week.

antaresViz : plotMap

type: title-slide

exercise :

  • For the date mentioned before represend the mix of production in Europe and represend the echanges between areas.

antaresProcessing

title: false type: second-title-slide

antaresProcessing

antaresProcessing

type: code-slide

Major functions

functions
Surplus, surplusClusters, surplusSectors Compute economic surplus
addDownwardMargin, addUpwardMargin Add downward and upward margins of areas
addExportAndImport Add export and import of areas or districts
addNetLoad Add net load of areas
compare Compare two antaresDataTable
Modulation Compute the modulation of cluster units

Upward margin

type: code-slide

exercise :

  • Compute the upward margin for the all simulations and represent it in a map for the dates mentioned before.

You can propose new features or report bugs on github

type: github title: no

You can propose new features or report bugs on github

Thanks

title: no type: thanks

Thanks you for your attention, don't forget R is great and you can call us anytime if you need help.