for this month's issue of our newsletter we have picked two papers about the operation of a 35 MW heat pump and about the process to develop business models and ownership structures for the construction of new district heating networks. Furthermore, we give a short overview of additional recent papers with topics ranging from uncertainty in transition roadmaps for existing district heating networks to synthetic data to train machine learning models for fault detection. We hope that you find these research inputs as inspiring for your work in the energy transition as we did.

Simulation and Experiment for Operation of a 35 MW Heat Pump

Large-scale heat pumps are widely considered to be a key technology for the transition of district heating towards a carbon-free future. And several projects are already underway to integrate large-scale heat pumps into existing networks as a replacement for former fossil-fired heat generators, or into new networks as a future-proof heat source. In this context, the very recent paper Dynamic Simulation and Experimental Validation of a 35 MW Heat Pump Based on a Transcritical CO2 Cycle by Wolscht et al. provides deep insights into the operation of such large-scale heat pumps. The paper describes a testbed setup for a heat pump unit with a maximum heat output of 35 MW at a COP of approximately 3.3 (requiring a maximum electricity input of 10.5 MW). Furthermore, the paper presents a detailed simulation model of the heat pump cycle written in the modeling language Modelica as well as a validation of the model based on measurements from the testbed setup.

The results of the paper show that the simulation model can match the measurement data with a high accuracy. If you're interested in the dynamic operation of large-scale heat pumps, the presented data includes many insightful details, especially for the dynamic operation of the heat pump. That includes a load change in which the electric power input of the heat pump is reduced from 100 % to 20 % in a time of only 30 - 40 seconds. Taking into account that this is a reduction from 10.5 MW to 2.1 MW, this example demonstrates the impressive flexibility of large-scale heat pumps. Especially from a sector-coupling-perspective this has important implications both for future operations of district heating networks and the electric grid. And furthermore, we think that this paper is a good indicator for the level of sophistication needed to optimally integrate large-scale heat pumps into future district heating.

Business Models and Ownership Structures for District Heating

As district heating is receiving a lot more attention in the context of the urgently needed energy transition, the scope of research to solve our current challenges goes far beyond technical aspects. Other important questions revolve around business models, social impact, and ownership structures. An interesting input to that discussion comes from the paper Sustainable deployment of energy efficient district heating: city business model by Pardo-Bosch et al. For the use case of a new-built district heating network in San Sebastián, the paper describes the process involving the municipality and other stakeholders to develop a business model for that network. This process lead to a structure in which the network is publicly owned and operated in a public-private partnership.

Throughout the paper, this process is documented with a detailed description of the roles and interests of different stakeholders within a Value Creation Framework, including e.g. citizens, the municipality as well as the district heating operator. We think that such a project description can be a helpful guide for other projects developing the foundations for new district heating networks. This is also highly relevant in the context of several initiatives across different countries. For example for the push towards municipal heat planning in Germany, where the municipality initiates a process to plan future heating solutions. And having a seamless continuation from this process towards the actual commissioning and operation of new district heating networks will be a key success factor. And in another example, there is already a preview available for an upcoming paper titled Why go public? Public configurations and the supportive and divergent views towards public district heating in the Netherlands by Herreras Martinez et al. which addresses related topics from the perspective of use cases in the Netherlands, further showing the international relevance of these topics.

Further Reading

In addition to the two papers we described in more detail above, we found several other papers recently published with very interesting topics and results. To also include those we recommend to look into the following publications, depending on your specific interests:

In the paper A profitability index for rural biomass district heating systems evaluation by Soltero et al., the authors show that while linear heat density is a good indicator for profitability of larger, high-density district heating networks. For rural networks based on a biomass heat source, the paper shows the shortcomings of such a linear heat density approach and instead suggests an alternative perspective based on analyzing the investments, operation, and biomass costs.

Regarding the application of machine learning algorithms to detect faults in district heating systems, a lack of training data is currently a major hurdle for progress. To overcome this problem, Vallee et al. describe the Generation and evaluation of a synthetic dataset to improve fault detection in district heating and cooling systems in a new paper. The results show, that models trained on such synthetic data perform very well on easier tasks and are also promising for more difficult fault detection tasks with the clear possibility to successfully transfer such pre-trained models into applications for real systems.

And we all know that uncertainty is a major challenge when planning transition roadmaps for district heating and cooling. Thus, the paper Roadmaps for heating and cooling system transitions seen through uncertainty and sensitivity analysis by Zhang et al. addresses an important and timely topic.

The next issue of our newsletter will be published on September 6, 2023. In the meantime, we invite you to have a look at our redesigned blog and follow us on LinkedIn.