Household energy consumption enriched with weather data in northeast of Mexico

Published: 25 January 2024| Version 1 | DOI: 10.17632/tvhygj8rgg.1
Contributors:
Baldemar Aguirre-Fraire,
,

Description

The dataset encompasses information obtained from energy consumption in a residence located in the northeast region of Mexico over a span of 14 months. Each data point was recorded at one-minute intervals and features energy-related metrics enriched with weather data provided by OpenWeather, under a free license. The variable descriptions are outlined as follows: date active_power current voltage reactive_power apparent_power power_factor main (categorical weather conditions) description (detailed categorical weather conditions) temp feels_like temp_min temp_max pressure humidity speed (wind speed) deg (wind degree) temp_t+1 (forecast of temperature for the next day) feels_like_t+1 (forecast of feels like for the next day) This dataset provides a comprehensive insight into the energy dynamics of the household, coupled with meteorological influences, offering a nuanced perspective on energy consumption patterns over time.

Files

Steps to reproduce

We used a smart meter device (AT-Q-SY1 WiFi) connected to the main energy source of the house. The data was transmitted to a Raspberry Pi through WiFi. We used a python script to organize and store the data. In addition, we connected to OpenWeather API to get weather data. Finally, we used python to integrate all data together.

Institutions

Universidad Autonoma de Coahuila

Categories

Time Series Prediction, Domestic Energy Consumption, Multivariate Model, Energy Forecasting, Applied Machine Learning

Funding

Consejo Nacional de Humanidades, Ciencias y Tecnologías

1243854

Licence