Skip to content

Latest commit

 

History

History
77 lines (45 loc) · 2.82 KB

code_overview.md

File metadata and controls

77 lines (45 loc) · 2.82 KB

RE-Europe code overview

This document attempts to give an overview of the scripts needed to be run in order to re-generate the dataset. For a list of required data, see Data_Overview.md.

This document is divided into 5 parts, each dealing with a separate part of the dataset:

  1. Transmission Network
  2. Generator Database
  3. Renewable production data
  4. Demand data
  5. Saving CSV files

In order to proceed with points 2-4, point 1 must have been executed. Point 4 further requires that the latitude/longitude grid from point 3 has been extracted. Point 5 depends on the relevant parts of the preceding points.

1. Transmission network

Network building/data extraction

Extract data from the Bialek network dataset into .csv files as described in Scripts/Network_latlon/README.md . Run Scripts/Network_latlon/1-Parse_network_data.py to parse the network data. Run Scripts/Network_latlon/2-Fit_positions.py to add positional data to the network.

A list of bus positions is located in Scripts/Network_latlon/coordinates-mercator.txt -- The user should check that the 'x' and 'y' columns of this datafile matches their extracted data.

2. Generator database

Generator extraction from Global Energy Observatory data

(Requires the network to have been built.) Download all relevant datafiles from www.globalenergyobservatory.com in kml format and place them in ./Data/Generator_Datafiles/excel_files. Run ./Scripts/Generator_Handling/1-Extract_GEO_Files to generate a database. Run ./Scripts/Generator_Handling/2-assign_generators_to_nodes to associate generators with network nodes, and build the final database.

3. Renewable production signals

Renewable forecast data conversion

TODO: Add me

Interpolating and converting real-time renewable production signal

TODO: Add me

Building Projection matrices for wind, solar and load

TODO: Add me

Aggregating wind and solar production to the nodal domain

TODO: Add me

Notes for converting COSMO data

TODO: Add me

4. Demand signal

ENTSOE load data to nodal signals

(Requires the network to have been built.)

(Requires the latitude and longitude grids from the ECMWF data to be present as ./Data/Metadata/{lats, lons}_ECMWF.npy)

(Requires the projection matrixes from the ECMWF data to have been built: ./Data/Metadata/loadtransfercsr_ECMWF.npz)

Running extract_excel_files.py parses the raw ENTSO-E country packages into .npy files.

Running load_to_maps.py then projects these .npy files according to the population in each grid cell.

Finally, running project_nodal_load.py then aggregates these maps of electrical demand onto the nodal domain.

5. Saving CSV files

The scripts located in ./Scripts/Save_{forecast,network,signal}_csv/ translates .npy files into the csv format, with the output located in ./Output_Data/{Nodal_FC,Metadata,Nodal_TS}/, respectively.