If you have any further questions regarding our services or see potential for cooperation, please contact us by email or phone number given below:
© heatbeat engineering GmbH & heatbeat nrw GmbH 2017-
Dear reader,
For the 41st issue of our heatbeat Research Newsletter, we have once again two recent research findings from the field of district heating and cooling. In our opinion, both articles highlight important trends and aspects that we need to consider more frequently in the design and operation of district heating networks in the future. The first article deals with the base load capacity of waste heat sources in district heating networks. The second article addresses the topic of predictive control and which algorithms can be used for this purpose.
The challenge of renewable energies is that they are often only available at varying times. In Denmark, for example, there are several heating networks that use solar thermal energy as a heat source, but their output fluctuates both daily (day/night) and throughout the year. Base load heat sources, such as waste heat and geothermal energy, will therefore play a decisive role in the future heat supply. We have already reported on this in several newsletters. The article presented here by D. Moreno et al. "Exploring the location and use of base load district heating supply. What can current heat sources tell us about future opportunities?" assesses how these heat sources can be used in district heating networks in the future. The methodology presented is applied to the example of Denmark.
A total of 360 district heating networks in Denmark were included in the analysis. A three-stage methodology is used for each of these district heating networks. First, the existing district heating networks are documented using a GIS analysis and possible expansion areas for the district heating network are determined on the basis of areal heat densities. The second step is the potential analysis, which takes into account industrial waste heat and geothermal energy in particular for the base load. Only sources which are located within 2 km of the district heating network area are considered. For the assessment of industrial waste heat waste heat, public data and existing methods are used and enhanced. The paper presents the waste heat potential for 46 industry branches. Finally, the base load capacity of the potential identified for the heating networks and their extensions is verified using annual duration curves.
The results show that in the future (thermal insulation + expansion of heating networks) a heat demand of around 30 TWh/a can be expected in heating networks in Denmark. Around 20 % of this can be covered by industrial waste heat sources capable of meeting base loads. Only waste heat sources above 60 °C were taken into account here, so that there is even higher potential in combination with heat pumps. Around half of the district heating systems examined have usable geothermal sources in the direct surroundings of the heating network.
Currently, district heating networks are often controlled with preset rules or manually. Both preset rules and manual control are often based on experience and can guarantee safe network operation. However, energy efficiency gains through temperature reduction or sector coupling to the electricity sector (use of heat pumps) are becoming increasingly important for the control of heating networks. The complexity increases with the number of different generators in the network and yet the output, temperature and pressure maintenance must be controlled in such a way that reliable operation is possible at all times. Predictive control optimization can be a valuable addition here. In the article "Mixed-integer non-linear model predictive control of district heating networks" by J. Jansen et al., two model predictive control are compared and applied in a simulation study.
The work by J. Jansen et al. compares linear and non-linear optimization approaches. Due to the large storage masses of the pipe system, the building energy demand and centralized or decentralized storage, a district heating network system is generally subject to a large number of dynamic processes and is therefore non-linear. MPC controllers are used to predict certain input data in the future (e.g. heat demand of the buildings, renewable energies) and use these parameters to determine the optimum control settings. The two approaches developed in the thesis are applied to a small heating network with a total of 12 buildings. The heating network is supplied by a heat pump (geothermal source) and a small share of solar thermal energy. Each building has a 950 liter buffer tank for domestic hot water, which provides the necessary temperature with a booster heat pump.
The results show that both approaches (linear and non-linear) lead to similar results in terms of control quality. A major advantage of the linear approach is a 7 - 9 times faster solution compared to the non-linear optimization. Nevertheless, the advantages of the non-linear approach are emphasized, which can be better generalized and requires less adjustment of the selected parameters than the linear approach (due to the better representation of real physics). We find both approaches very interesting and believe that MPC will play a decisive role in the future control of heating networks.
In addition to the two papers mentioned above, we can recommend the article "District heating load patterns and short-term forecasting for buildings and city level" by P. Hua et al., which presents methods for predicting demand in buildings of a district heating network, which can be used in particular for the application of MPC controllers.
The next issue of our newsletter will be published on April 3, 2024.