Notice: The web app service has been disabled due to server cost reasons. However, you can request synthetic heat pump load profiles for free by contacting leo.semmelmann [at] kit.edu.
City for underlying temperature profile:
Köln-Stammheim
To which email address should we send the generated synthetic heat pump profiles?
Synthetic Heat Pump Load Profiles
Heatpump.ninja enables the synthetic generation of heat pump load profiles for different geographies and respective temperature profiles for research in energy informatics. The data underlying the project is based on the dataset of Schlemminger et al. [1], who contributed a high-quality dataset of water-to-water heat pump profiles from Hamelin, Northern Germany. The synthetic heat pump profile creation is based on a K-Means clustering approach, described in [2]. The underlying source code is available open-source on Github. A ready-made dataset of 500 synthetic residential heat pump load profiles for Karlsruhe can be downloaded on Zenodo.
[1] Marlon Schlemminger et al. “Dataset on electrical single-family house and heat pump load profiles in Germany”. In: Scientific Data 9.1 (2022), p. 56.
[2] Semmelmann et al. “Generating synthetic load profiles of residential heat pumps: a k-means clustering approach”. In: Energy Informatics (2023).
Limitations
We note that our results are especially valid for countries with building structures and temperature profiles like in Germany. Also, we note that results might differ for alternative heat pump types, such as the most proliferated air-to-air heat pumps. However, due to comparable coefficients of performance, we argue that our synthetic heat pump load profiles can give a good indicator of relevant heat pump loads and are hence suitable for energy system modeling and simulations. Furthermore, the underlying households are equipped with thermal solar collectors that provide domestic hot water during summer months and hence reduce the heat pump load during them. Although this has to be taken into account when analyzing synthetically generated heat pump loads during summer months, we argue that the thermal collectors do not interfere with the overall applicability of the profiles, since the largest part of heat pump loads is a) caused by heating and b) concentrated at winter months, for which both the thermal collectors do not play a role.
Contributors
Leo Semmelmann, Patrick Jarquart, Armin Golla
Questions about the project?
In case you haven't received your synthetic profiles yet, please look in your spam folder. At times of multiple requests, it may take a while until you receive your synthetic load profiles.
If you are interested in creating larger number of synthetic heat pump load profiles (or for different cities) feel free to get in touch with leo.semmelmann [at] kit.edu.