K-means cluster analysis of hourly measured power demand in district heating


  • Uwe Radtke Hungarian University of Agriculture and Life Sciences, Doctoral School in Management and Organizational Sciences, Kaposvár Campus




smart meter data analysis, carbon emissions reduction, waste heat utilization, energy efficiency, sustainable energy


Within this paper the actual cluster analysis is performed in order to identify the clusters within the Kaposvár district heating. The data was measured not directly at the households but at the heat transfer stations. The smart meters were installed at the heat transfer stations for several reasons, not to measure and control the needed supply temperature but also to identify leakages quicker and easier. The method used is k-means which at the end did determine four different clusters: high demand at low operating hours, high demand at long operating hours as well as low demand at high operating hours and low demand for low operating hours. The details and the determined values can be used for further research and already first further steps towards identifying new heat sources.


Andrews, Dave. 2009. “"Carbon Footprints of Various Sources of Heat - Biomass Combustion and CHPDH Comes Out Lowest " - William Orchard.” Claverton Group, February 7, 2009. https://claverton-energy.com/carbon-footprints-of-various-sources-of-heat-chpdh-comes-out-lowest.html.

Bánkuti, G Zanatyné Uitz, S. 2021. “A Good Practice in Urban Energetics in a Hungarian Small Town, Kaposvar.” In Handbook of Smart Energy Systems, edited by Mahdi Fathi, Enrico Zio, and Panos M. Pardalos, forthcomming. Cham: Springer International Publishing; Imprint: Springer.

Csima, F., & Szendefy, J. 2009. “Synergic Consideration of Competitiveness and Ecology in Biogas Production and Use.” Regional and Business Studies 1 (1): 45–48.

Glass online. 2022. “Sisecam to Build Glass Packaging Plant in Kaposvár, Hungary | GlassOnline.Com - the World's Leading Glass Industry Website.” Accessed October 03, 2022. https://www.glassonline.com/sisecam-to-build-glass-packaging-plant-in-kaposvar-hungary/.

IEA. 2022. “District Heat Production by Region, 2020, and World Average Carbon Intensities of District Heat Supply in the Net Zero Scenario, 2020-2030.” Accessed April 10, 2022. https://www.iea.org/reports/district-heating.

Jain, Anil K. 2010. “Data Clustering: 50 Years Beyond K-Means.” Pattern Recognition Letters 31 (8): 651–66. https://doi.org/10.1016/j.patrec.2009.09.011.

Kaposvári Municipal Property Management and Service Co., Ltd. 2022. “Green Heating Plant - the Freedom of District Heating Reaches Apartments.” Accessed September 18, 2022. https://www.kaposho.hu/hirek/zoldfutomu-a-lakasokig-er-a-tavhoszabadsag.html.

MacQueen, J. B. 1967. “Some Methods for Classification and Analysis of Multivariate Observations.” Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability 1: 281–97. https://projecteuclid.org/Proceedings/berkeley-symposium-on-mathematical-statistics-and-probability/proceedings-of-the-fifth-berkeley-symposium-on-mathematical-statistics-and/toc/bsmsp/1200512974.

Radtke, Uwe. 2022. “A Brief Literature Review of Structuring District Heating Data Based on Measured Values.” (122(2)): FORTHCOMING.

Ramos, Sérgio, João M. Duarte, F. Jorge Duarte, and Zita Vale. 2015. “A Data-Mining-Based Methodology to Support MV Electricity Customers’ Characterization.” Energy and Buildings 91:16–25. https://doi.org/10.1016/j.enbuild.2015.01.035.

Tureczek, Alexander Martin, Per Sieverts Nielsen, Henrik Madsen, and Adam Brun. 2019. “Clustering District Heat Exchange Stations Using Smart Meter Consumption Data.” Energy and Buildings 182:144–58. https://doi.org/10.1016/j.enbuild.2018.10.009.

Wikipedia. 2022. “K-Means Clustering.” Accessed September 18, 2022. https://en.wikipedia.org/wiki/K-means_clustering#cite_note-7.




How to Cite

K-means cluster analysis of hourly measured power demand in district heating. (2022). REGIONAL AND BUSINESS STUDIES, 14(1), 57-72. https://doi.org/10.33568/rbs.3343