THE GENERATION BATTLE – 4th versus 5th generation of District Heating

In recent years, the 5th generation of DH has been emerging. But what are the drivers behind the development, and how is it positioned compared to the prior generations? Before a meaningful answer, one needs to know how the 5th generation’s definition deviates from the definition of the 4^th generation and what impact the difference has on the supply system.

By Oddgeir Gudmundsson and Jan Eric Thorsen, Danfoss Climate Solution, Denmark

District heating (DH) is here to stay. Looking back on the history of DH, it goes quite some years back. Over the years, it has developed to fulfill the demands as they came up, typically driven by the need for reduced investment and heat costs, low er equipment space demands, concerns of energy efficiency, environment, longer lifetime, and lower fire risks. In 2014 an article [1] was published that categorized the historical development into four generations. Each generation was defined by significant technological or purpose changes compared to the prior generation. Currently, most DH schemes being operated are categorized as being at the 3rd generation DH technology stage and starting their transition to the 4th generation.

The transition is driven by the challenge of the future non-fossil and renewable-based energy system. At the same time, active work is being performed in the research community to see how this great technology can be further innovated. This has led to the concept of an ambient temperature distribution grid, commonly called the 5th generation DH. Each end user operates his heat pump to adapt the supply temperature to his needs. The research has mainly been focused on case studies and a few small-scale concept validation projects. What has generally been missing is a direct comparison of the 5th generation to the 4th generation. What are the differences between the generations that impact the integration potential of the energy supply system? This article tries to address that.

“The fifth generation”

In recent years, the fifth generation of DH has been emerging. But what are the drivers behind the development, and how is it positioned compared to the prior generations? Before a meaningful answer, one needs to know how the fifth generation’s definition deviates from the definition of the 4th  generation and what impact the difference has on the supply system.  To answer this, one needs to know what the original idea was behind the definition of the 4th generation. This can be read from the generation figure, Figure 1, which shows the transition from fossil fuels to a renewable energy source; it further shows an intelligent integrated energy system where the heating sector is coupled with the power, industry, and cooling sectors. It is about energy efficiency, it is about sustainability, and it is about demand-driven energy systems.  When looking into the main ideas behind the 5th generation, one quickly finds that it is identical to the 4th generation. It is about energy efficiency, sustainability, and demand-driven thermal supply. The difference between the generations does not lie in the purpose of the system but in how the systems are designed to fulfill their purpose.   

Before digging into the differences, we need to define what energy sources and heat plants mean:

Figure 1: Illustration of the concept of 4th Generation DH compared to the previous three generations. 

Energy source: The energy source is the primary energy input to the system. It can be any energy source before the conversion to the desired energy form. In the case of a building, thermal demands are heated at an immediately functional temperature level.

Heat plant: Heat plant is the conversion technology used to convert the energy source to the desired temperature level.

With these definitions in place, we can make a clear distinction between the 4th generation and the 5th generation:

• The 4th and 1st to 3rd generations rely on centralized heat plants to convert the energy source to heat at immediately useful temperature levels for the end-user.
• The 5th generation is taking a different approach: to deliver the energy source to the end-user, which uses his heat plant to convert that energy source to heat at an immediately useful temperature level.

For the 4th generation, the form/type of the applied primary energy source and central heat plants are not restricted. For the 5th generation, the primary energy source is low-quality heat at too low levels for direct utilization, mainly ambient heat sources. The heating plant is a building-level / end-user heat pump.

This individualization of the temperature upgrading of ambient sources provides the fundamental difference between the 5th generation and the prior generations. It provides the platform to compare the 5th generation to the 4th generation.

From a technical point of view, the moving of the heating plant from a central location towards the end-user has some benefits, for example:

• Distribution heat losses may become insignificant, even irrelevant, as no energy has been spent upgrading the input energy, ambient heat, to useful temperature levels.
• With no distribution losses, pipe insulation becomes irrelevant, which in theory, will lead to a cheaper distribution network.
• The heat plants (end-user heat pumps) can be adapted to the temperature requirement of each end-user, in theory leading to better heat generation efficiency.
• Possibility to integrate cooling into the same system by enabling the end-user heat pumps to operate in a cooling mode and deliver the waste heat into the distribution grid.

From the outset, these are quite promising benefits, at least at first glance. Since everything is subjected to local conditions, and even from the viewer’s standpoint, it is perhaps a good idea to look into these benefits from a broader perspective.

5G claim: No distribution loss is the best.  

First, nobody likes losses, except for parents losing games to their happy and smiling kids (which is technically a win in the bigger picture). Now, if no one likes losses, why should we bother with a closer look at the lossless distribution? If there are winners, there are also losers. We just need to ensure that the losers are neither the end-users nor the society.  

As the saying goes: You lose some, and you win some. The winnings from having no distribution loss are that no heat is lost between the heat source and the end-users, and there is no need for insulation. No insulation further means a cheaper distribution system (at least in theory). Potential losses from having “no losses” could, on the other hand, be lost opportunities and secondary inefficiencies.  

As lost opportunities can be too many to count, one should   focus on those who are likely to be most important, such as:  

• Ability to use alternative energy sources capable of generating heat at immediately useful temperature levels, such as waste heat from industry, power production, or any other new types of energy sources that may be discovered in the local area  

• • Operating an uninsulated pipe network at higher than the ambient temperature would lead to significant distribution losses.  

• Ability to use more energy and cost-efficient large-scale centralized heat plants 
• Avoidance of upgrading or specifying unnecessarily strong power distribution grid 

• • As thermal demand is a vast majority of building energy demands, there is a high risk that the heating demand in an individualized electrified heating system would define the power connection capacities.  

• Ability to have long-term decoupling of the thermal demand and thermal generation  

• • Long-term decoupling can only be achieved via central thermal storage capable of fulfilling thermal demands at immediate functional temperature levels  

5G claim: Uninsulated pipe network is cheaper.  

While saving investment costs by using an uninsulated pipe is certainly an interesting point. It also deserves a closer look:  

One needs to consider the system design conditions, particularly the expected temperature difference between the supply and return flow, as this will define the required pipe diameters of the distribution network. In that respect, two main factors can negatively influence the design of ambient loop systems.  

• The ambient temperature heat sources tend to cool down during the heating season. 
• The higher the heat pump efficiency, the smaller the temperature difference is across the evaporator, e.g., the cooling of the heat source.  

Pipe dimensions for capacities below 2,000 kW

Pipe dimensions for capacities above 2,000 kW

Figure 2: Impact of the temperature difference on the pipe dimension for a given heat transfer capacity.

Both points lead to squeezing the system temperature difference and lead to the need for large pipe dimensions. From the perspective of potential cooling demands being used for regenerating the heat supply in the 5G system, it should be considered that the likelihood of any user operating in a cooling mode is particularly low during the heating season. The lack of regeneration from cooling leads to the pipe network being dimensioned like traditional district heating systems, from the heat source to the end-users.  

The impact of the temperature difference on the required pipe size for capacities of a few kilowatts to 10 megawatts is shown in Figure 2.  

Not particularly surprisingly, the figure shows that the lower the temperature difference, the larger the required pipe dimension is. This is important as the pipe dimension directly impacts the trenching cost of the network. The bigger the pipe, the bigger the trench. Secondly, the bigger the pipe is, the more costly it becomes.  

We have an additional parameter when comparing the 5th generation with the 4th generation. For dimensions below DN 200, the insulated pipe system can be optimized by incorporating both the supply and the return pipe into the same sleeve, e.g., Twin pipes. The twin-pipe concept has the benefit of reducing the trench requirement compared to two single pipes; see the right side of Figure 3. Additionally, twin pipes significantly reduce heat losses from the distribution pipe network compared to a set of single pipes. This boils down to the following fact: The generally larger pipe dimensions required for the 5th generation and the availability of the Twin pipe concept for the 4th generation reduce the cost-benefit of an uninsulated pipe network.  

All in all, these factors can effectively neutralize the cost-benefit of applying an uninsulated pipe network.  

Two single uninsulated pipes in a trench

Twin pipe in a trench

Figure 3: Trenches of uninsulated pipes, top, and twin pipes, bottom.

5G claim: End-user heat generation leads to higher efficiencies. 

This is a fascinating point. Without a doubt, the efficiency of heat pumps is higher the lower the temperature lift, i.e., the less you need to increase the supply temperature, the more efficient your operation becomes. 

There is, however, more at stake here than one might initially consider. For example: 

1. Centralized heat generation enables economy of scale by taking advantage of simultaneities and the aggregation of  the heat demand
2. Large-scale centralized heat pumps are both more cost and energy efficient than small end-user heat pumps
3. Large-scale centralized heat pumps are professionally maintained, and the operation is optimized, leading to stable and energy-efficient operation and long lifetimes.
4. Centralized heat pumps connect to a higher voltage grid, which avoids potential power losses in the power transformation stations and distribution grid.   
5. In principle, the savings of losses in the power grid can go a long way towards the heat loss in the 4th generation district heating system, considering the COP of the centralized heat pump.
6. Centralized heat pumps can access power at a lower cost due to the aggregated demand and by connecting with the high-voltage grid.
7. In combination with a district cooling system, both sides of the heat pump can be exploited, creating a uniquely high  system energy efficiency.
8. While this is commonly mentioned as an inherent part of the 5G systems, it is not to the same extent, as waste heat from the cooling operation in 5G systems is not delivered  at a useful temperature level to the loop; it is mainly taking over part of the loop regeneration purposes of the heat source.
9. Centralized heat pumps and centralized thermal storage are uniquely positioned to decouple the heat/cool demand and the heat/cool generation, enabling it to take greater advantage of electricity tariffs and low carbon power periods and provide power grid balancing services. 

Comparison of the solutions

The annual cost of heat for the average connected building in Copenhagen

Figure 4: Levelized cost of heating for high energy (HE) and low energy (LE) buildings in Copenhagen for various thermal source temperatures. ATDH is the 5th generation, and LTDH is the 4th generation.

The annual cost of heat for the average connected building in London

Figure 5: Levelized cost of heating for high energy (HE) and low energy (LE) buildings in London for various thermal source temperatures. ATDH is the 5th generation, and LTDH is the 4th generation.

The above points translate to better cost optimization of heat generation.

But there is also the fact that if there is an insulated network, the system is significantly better positioned to enable energy-efficient utilization of any heat source with a higher temperature level than the ambient. In ambient loops, higher waste heat temperatures than the ambient could only be used with high heat losses in the uninsulated network.

5G claim: Cooling is integral to the 5G solution and effectively separates 5G from 4G.

This is the only part where the 5th generation has a clear advantage over the 4th generation regarding residential cooling demands, at least compared to the original definition of the 4th generation, written by Scandinavians, where residential cooling demand is virtually non-existing.

In principle, the cooling demand could be integrated into the 4th generation in the same way as in the 5th generation by using the distribution network as a heat sink for end-user-located heat pumps.

For commercial cooling demands, which tend to be concentrated and more prominent than residential cooling demands, the most efficient supply system would be a dedicated district cooling system, as is applied in Paris, Stockholm, Helsinki, Copenhagen, and many more cities around the world.

How does the 5th generation economically stack up to the 4th generation?

Due to the relatively few 5th-generation systems, there is a general lack of economic comparison of these solutions. This has, however, been compared in the paper “Economic comparison of 4GDH and 5GDH systems – Using a case study” [2]. The case analysis compared the cost of these two generations supplying a suburban area in two locations with different climates, Copenhagen and London. The case analysis further investigated the impact on the system economics if it would be supplying either existing high-energy buildings or new low-energy buildings. Finally, the case analysis considered the impact the thermal source’s temperature would have on the economics of the system. The results of the case analysis are shown in Figures 4 and 5. The figures clearly show that 4th-generation district heating systems have significantly better economics than 5th-generation systems.

Final words

As the case study focused on a suburban setting, typically considered outside of the core district heating zones, densely populated inner cities with concentrated heating demands, it points to that the 5th generation requires specific conditions to be in place to ensure its competitiveness. These conditions are like to occur when an economy of scale from central heat generation is impossible. These conditions might occur in small settlements with relatively few houses. Instead of each house operating its own ambient heat source, for example, a geothermal loop, the settlement could share a larger and more cost-efficient geothermal loop. In that sense, the 5th generation can also be classified as an extended individual heat pump system.

 

References

Lund, H. et al., 4th Generation District Heating (4GDH). Integrating Smart Thermal Grids into Future Sustainable Energy Systems. Energy, vol. 68, 1-11, 2014. https://doi.org/10.1016/j.energy.2014.02.089

Gudmundsson, O. et al., Economic comparison of 4GDH and 5GDH systems – Using a case study. Energy, vol. 238, 2021. https://doi.org/10.1016/j.energy. 2021.121613

For further information, please contact: Oddgeir Gudmundsson at og@danfoss.com