Electricity consumption dataset of a local energy cooperative.

Real-world data collections are generally not easily available. Energy measurements from buildings, houses and other devices can be used within different areas of research while being employed to plan or train models, allowing the improvement of power grid energy efficiency or providing more insight on how an energy community can work. This paper provides a dataset concerning a Portuguese community of 172 households that are geographically close to each other, enabling the establishment of relationships among buildings and the analysis of a community's power consumption. In addition to the consumed energy values, the related local weather information is included in the data. The intersection of weather data and energy measurements can be helpful to train AI models, contributing to explain variations in energy consumption and the absolute values of the energy readings. The inclusion of these weather parameters aims to unveil features that can correlate to the energy measurements, enabling them to be used in multiple areas of research. Hence, it will provide added value to the data as it can be reused to explore Machine Learning algorithms or community energy planning by grid operators.