Portuguese Transmission System Aggregated Load Time Series and Encoded Time Covariates
Aggregated load time series of the Portuguese TSO (Transmission System Operator) for 2018 and 2019 (15 minute resolution) accompanied by encoded (in a cyclical manner) time covariates. Useful for TSO demand forecasting experimentation.
File 1: Load Series
- Features: Datetime, Load Value
- Resolution: 15 min
- Years: 2018, 2019
File 2: Time Covariates
- Features: Datetime, year, month_sin, month_cos, dayofyear_sin, dayofyear_cos, hour_sin, hour_cos, minute_sin, minute_cos, dayofweek_sin, dayofweek_cos, weekofyear_sin, weekofyear_cos, holidays
- Resolution: 15 min
- Years: 2018, 2019
Developed by EPU-NTUA in the context of I-NERGY project.
The zip file includes two csv files at 15 minute resolution. The first one contains the aggregated load values as recorded, provided and owned by the Portuguese TSO (REN). The second one contains encoded time covariates that can be fed as input into machine learning and deep learning models.
This is a contribution of the I-NERGY project for AI4EU.
DSS Lab does not claim ownership of the load data as they are publically available by REN. It only provides them accompanied by encoded time covariates in order to enable easy experimentation with ML models for data scientists.