XAI4HEAT SCADA Dataset 2024

Real-world heating data were extracted from the SCADA system of a local District Heating System (DHS) at the Faculty of Mechanical Engineering in Niš, Serbia. The dataset includes one-hour resolution parameters from five substations in residential buildings over five heating seasons, capturing temperatures of the heating fluid in supply and return lines, energy transmission metrics, and outdoor temperatures. It enables research on explainable intelligent control of DHS, focusing on heat demand response and load shifting to optimize performance and sustainability. Additionally, it enhances transparency in urban heat consumption through Explainable AI, offering interpretable insights into heat supply variations and facilitating what-if analyses for stakeholders.