GoiEner dataset processing files
This repository contains scripts for processing and cleaning GoiEner's smart meter data. It contains three R scripts named simel_to_user.R, user_to_raw.R and raw_to_clean.R. These scripts convert SIMEL files to USER files, USER files to RAW files (raw electricity usage CSV files), and RAW files to CLEAN files (clean electricity usage CSV files before being split into pre-, during, and post-COVID-19 files), respectively.
- R software and environment
-
simel_to_user.R: This script transforms SIMEL files from the directory indicated by the variableg_inputinto USER files in the directory indicated by the variableg_output. -
user_to_raw.R: This script takes all USER files from the directory indicated by the variablegoiener_users_folder, processes them, and transforms them into RAW files. The resulting RAW files are saved in the directory indicated by the variablegoiener_output_folder. -
raw_to_clean.R: This script transforms RAW files from the directory indicated by the variableinput_folderinto CLEAN files. The CLEAN files are stored in the directory indicated by the variableoutput_folder.
-
From SIMEL files to USER files (CUPS files): The
simel_to_user.Rscript collects entries related to each anonymized CUPS from all SIMEL files, and generates individual USER files for each CUPS. -
From USER files (CUPS files) to RAW files: The
user_to_raw.Rscript eliminates duplicate entries in each USER file that have the same timestamp. This step ensures the data is accurate and consistent.In addition, it processes the USER files to generate the RAW files in CSV format, extracting and organizing relevant information. -
Data Cleaning (From RAW files to CLEAN files): The
raw_to_clean.Rscript cleans the raw data in the RAW files and transforms them into CLEAN files. It imputes missing data, adjusts timestamps to local time, avoids COVID-19 lockdown dates, excludes short load profiles, and excludes 0-valued load profiles.
The original SIMEL files provided by GoiEner are available on Zenodo in the "GoiEner smart meters raw data" repository. The processed data are also available in the "GoiEner smart meters data" repository on Zenodo.
For any queries or issues, please contact carlos.quesada@deusto.es