District heating

District heating (also known as heat networks) is a system for distributing heat generated in a centralized location through a system of insulated pipes for residential and commercial heating requirements such as space heating and water heating. The heat is often obtained from a cogeneration plant burning fossil fuels or biomass, but heat-only boiler stations, geothermal heating, heat pumps and central solar heating are also used, as well as heat waste from factories and nuclear power electricity generation. District heating plants can provide higher efficiencies and better pollution control than localized boilers. According to some research, district heating with combined heat and power (CHPDH) is the cheapest method of cutting carbon emissions, and has one of the lowest carbon footprints of all fossil generation plants.^[1]

District heating is ranked number 27 in Project Drawdown's 100 solutions to global warming.^[2]^[3]

History

District heating traces its roots to the hot water-heated baths and greenhouses of ancient times, perhaps today most known in the Roman Empire. A hot water distribution system in Chaudes-Aigues in France is generally regarded as the first real district heating system. It used geothermal energy to provide heat for about 30 houses and started operation in the 14th century.^[4]

The U.S. Naval Academy in Annapolis began steam district heating service in 1853.^{[citation needed]} MIT began coal-fired steam district heating in 1916 when it moved to Cambridge, Massachusetts.^[5]^[6]

Although these and numerous other systems have operated over the centuries, the first commercially successful district heating system was launched in Lockport, New York, in 1877 by American hydraulic engineer Birdsill Holly, considered the founder of modern district heating.

Generations of district heating

Generally, all modern district heating systems are demand driven, meaning that the heat supplier reacts to the demand from the consumers and ensures that there is sufficient temperature and water pressure to deliver the demanded heat to the users. Each generation has a defining feature that sets it apart from the prior generations. The feature of each generation can be used to give an indication of the development status of an existing district heating system.

First generation

The first generation was a steam-based system fueled by coal and was first introduced in the US in the 1880s and became popular in some European countries, too. It was state of the art until the 1930s. These systems piped very high-temperature steam through concrete ducts, and were therefore not very efficient, reliable, or safe. Nowadays, this generation is technologically outdated. However, some of these systems are still in use, for example in New York or Paris. Other systems originally built have subsequently been upgraded.^[7]

Second generation

The second generation was developed in the 1930s and was built until the 1970s. It burned coal and oil, and the energy was transmitted through pressurized hot water as the heat carrier. The systems usually had supply temperatures above 100 °C, and used water pipes in concrete ducts, mostly assembled on site, and heavy equipment. A main reason for these systems was the primary energy savings, which arose from using combined heat and power plants. While also used in other countries, typical systems of this generation were the Soviet-style district heating systems that were built after World War II in several countries in Eastern Europe.^[7]

Third generation

In the 1970s the third generation was developed and was subsequently used in most of the following systems all over the world. This generation is also called the "Scandinavian district heating technology" because many of the district heating component manufacturers are based in Scandinavia. The third generation uses prefabricated, pre-insulated pipes, which are directly buried into the ground, and operates with lower temperatures, usually below 100 °C. A primary motivation for building these systems was security of supply by improving the energy efficiency after the two oil crises led to disruption of the oil supply. Therefore, those systems usually used coal, biomass and waste as energy sources, in preference to oil. In some systems, geothermal energy and solar energy are also used in the energy mix.^[7] For example, Paris has been using geothermal heating from a 55–70 °C source 1–2 km below the surface for domestic heating since the 1970s.^[8] Especially in the former Eastern Bloc nuclear energy has been used for district heating^[9]^[10] and new systems keep being installed in China.^[11] The source of the heat of nuclear district heating is virtually always waste heat from power reactors, but proposals to build dedicated heating reactors or to use the waste heat from repurposed spent fuel pools^[12] have been brought forth.^[13]^[14]

Fourth generation

The fourth generation is being developed,^[7] with the transition to the fourth generation already in process in Denmark.^[15]^[16]^[17] The fourth generation is designed to combat climate change and integrate high shares of variable renewable energy into the district heating by providing high flexibility to the electricity system.^[7]

According to the review by Lund et al.^[7] those systems have to have the following abilities:

"Ability to supply low-temperature district heating for space heating and domestic hot water (DHW) to existing buildings, energy-renovated existing buildings and new low-energy buildings."
"Ability to distribute heat in networks with low grid losses."
"Ability to recycle heat from low-temperature sources and integrate renewable heat sources such as solar and geothermal heat."
"Ability to be an integrated part of smart energy systems (i.e. integrated smart electricity, gas, fluid and thermal grids) including being an integrated part of 4th Generation District Cooling systems."
"Ability to ensure suitable planning, cost and motivation structures in relation to the operation as well as to strategic investments related to the transformation into future sustainable energy systems".

Compared to the previous generations the temperature levels have been reduced to increase the energy efficiency of the system, with supply side temperatures of 70 °C and lower. Potential heat sources are waste heat from industry, CHP plants burning waste, biomass power plants, geothermal and solar thermal energy (central solar heating), large scale heat pumps, waste heat from cooling purposes and data centers and other sustainable energy sources. With those energy sources and large scale thermal energy storage, including seasonal thermal energy storage, fourth-generation district heating systems are expected to provide flexibility for balancing wind and solar power generation, for example by using heat pumps to integrate surplus electric power as heat when there is much wind energy or providing electricity from biomass plants when back-up power is needed.^[7] Therefore, large scale heat pumps are regarded as a key technology for smart energy systems with high shares of renewable energy up to 100% and advanced fourth-generation district heating systems.^[18]^[7]^[19]

Fifth generation/cold district heating

A fifth-generation district heating and cooling network (5GDHC),^[20] also called cold district heating, distributes heat at near ambient ground temperature: this in principle minimizes heat losses to the ground and reduces the need for extensive insulation. Each building on the network uses a heat pump in its own plant room to extract heat from the ambient circuit when it needs heat, and uses the same heat pump in reverse to reject heat when it needs cooling. In periods of simultaneous cooling and heating demands this allows waste heat from cooling to be used in heat pumps at those buildings which need heating.^[21] The overall temperature within the ambient circuit is preferably controlled by heat exchange with an aquifer or another low temperature water source to remain within a temperature range from 10 °C to 25 °C.

While network piping for ambient ground temperature networks is less expensive to install per pipe diameter than in earlier generations, as it does not need the same degree of insulation for the piping circuits, it has to be kept in mind that the lower temperature difference of the pipe network leads to significantly larger pipe diameters than in prior generations. Due to the requirement of each connected building in the fifth-generation district heating and cooling systems to have their own heat pump the system can be used as both a heat source and a heat sink for the heat pump, depending on if it is operated in heating or cooling mode. As with prior generations the pipe network is an infrastructure that in principle provides open access for various low-temperature heat sources, such as ambient heat, ambient water from rivers, lakes, sea, or lagoons, and waste heat from industrial or commercial sources.^[22]

Based on the above description it is clear that there is a fundamental difference between the 5GDHC and the prior generations of district heating, particularly in the individualization of the heat generation. This critical system has a significant impact when comparing the efficiencies between the different generations, as the individualization of the heat generation moves the comparison from being a simple distribution system efficiency comparison to a supply system efficiency comparison, where both the heat generation efficiency as well as the distribution system efficiency needs to be included.

A modern building with a low-temperature internal heat distribution system can install an efficient heat pump delivering heat output at 45 °C. An older building with a higher-temperature internal distribution system, e.g. using radiators, will require a high-temperature heat pump to deliver heat output.

A larger example of a fifth-generation heating and cooling grid is Mijnwater in Heerlen, the Netherlands.^[23]^[24] In this case the distinguishing feature is unique access to an abandoned water-filled coal mine within the city boundary that provides a stable heat source for the system.

A fifth-generation network ("Balanced Energy Network", BEN) was installed in 2016 at two large buildings of the London South Bank University as a research and development project.^[25]^[26]

Heat sources

District heating networks exploit various energy sources, sometimes indirectly through multipurpose infrastructure such as combined heat and power plants (CHP, also called co-generation).

Combustion of fossil or renewable fuels

The most used energy source for district heating is the burning of hydrocarbons. As the supply of renewable fuels is insufficient, the fossil fuels coal and gas are massively used for district heating.^[27] This burning of fossil hydrocarbons usually contributes to climate change, as the use of systems to capture and store the CO₂ instead of releasing it into the atmosphere is rare.

In the case of a cogeneration plant, the heat output is typically sized to meet half of the peak winter heat load, but over the year will provide 90% of the heat supplied. Much of the heat produced in summer will generally be wasted. The boiler capacity will be able to meet the entire heat demand unaided and can cover for breakdowns in the cogeneration plant. It is not economic to size the cogeneration plant alone to be able to meet the full heat load. In the New York City steam system, that is around 2.5 GW.^[28]^[29] Germany has the largest amount of CHP in Europe.^[30]

A simple thermal power station can be 20–35% efficient,^[31] whereas a more advanced facility with the ability to recover waste heat can reach total energy efficiency of nearly 80%.^[31] Some may approach 100% based on the lower heating value by condensing the flue gas as well.^[32]

Nuclear fission

Around 40 nuclear power plants supply district heating, mostly in Russia, China and Eastern Europe.^[33] The heat produced by nuclear chain reactions can be injected into district heating networks. This does not contaminate the district pipes with radioactive elements, as the heat is transferred to the network through heat exchangers.^[34] It is not technically necessary for the nuclear reactor to be very close to the district heating network, as heat can be transported over significant distances (exceeding 200 km) with affordable losses, using insulated pipes.^[35]^{[clarification needed]}

Since nuclear reactors do not significantly contribute to either air pollution or global warming, they can be an advantageous alternative to the combustion of fossil hydrocarbons. However, only a small minority of the nuclear reactors currently in operation around the world are connected to a district heating network. These reactors are in Bulgaria, China, Hungary, Romania, Russia, Slovakia, Slovenia, Switzerland and Ukraine.^[36]^[33]

The Sibirskaya Nuclear Power Plant in USSR was the first nuclear CHP plant, supplying district heating to Seversk since 1961 and to a part of Tomsk since 1973, stopped in 2008.^[37] The Ågesta Nuclear Power Plant in Sweden was an early example of nuclear cogeneration, providing small quantities of both heat and electricity to a suburb of the country's capital between 1964 and 1974. The Beznau Nuclear Power Plant in Switzerland has been generating electricity since 1969 and supplying district heating since 1984. The Haiyang Nuclear Power Plant in China started operating in 2018 and started supplying small scale heat to the Haiyang city area in 2020. By November 2022, the plant used 345 MW-thermal effect to heat 200,000 homes, replacing 12 coal heating plants.^[38]

Recent years have seen renewed interest in small modular reactors (SMRs) and their potential to supply district heating.^[39] Speaking on the Energy Impact Center (EIC) podcast Titans of Nuclear, Christer Dahlgren, principal engineer at GE Vernova Hitachi Nuclear Energy, noted that district heating could be the impetus for the construction of new nuclear power plants in the future.^[40] EIC's own open-source SMR blueprint design, OPEN100, could be incorporated into a district heating system.^[41]

Natural underground heat

History

Geothermal district heating was used in Pompeii, and in Chaudes-Aigues since the 14th century.^[42]

Denmark

Denmark has one geothermal plant in operation in Thisted since 1984. Two other plants are now closed, located in Copenhagen (2005–2019), and Sønderborg (2013–2018). Both suffered issues with fine sand and blockages.^[43]^[44]^[45]

The country's first large-scale plant started in Skejby in Aarhus in 2025,^[46] and by the end of 2030, it is expected to be able to cover approximately 20% of the district heating demand in Aarhus.^[47]

Iceland

; District heating § National variation

Most of Iceland's population is heated by geothermal heat.

United States

Direct-use geothermal district heating systems, which tap geothermal reservoirs and distribute the hot water to multiple buildings for a variety of uses, have existed for over a century in the US but are uncommon.

In 1890, the first wells were drilled to access a hot water resource outside of Boise, Idaho. In 1892, after routing the water to homes and businesses in the area via a wooden pipeline, the first geothermal district heating system was created.

As of a 2007 study,^[48] there were 22 geothermal district heating systems (GDHS) in the United States. As of 2010, two of those systems have shut down.^[49] The table below describes the 20 GDHS currently^[when?] operational in the US.

System name	City	State	Startup year	Number of customers	Capacity (MWt)	Annual energy generated (GWh)	System temperature
System name	City	State	Startup year	Number of customers	Capacity (MWt)	Annual energy generated (GWh)	°F	°C
Warm Springs Water District	Boise	ID	1892	275	3.6	8.8	175	79
Oregon Institute of Technology	Klamath Falls	OR	1964	1	6.2	13.7	192	89
Midland	Midland	SD	1969	12	0.09	0.2	152	67
College of Southern Idaho	Twin Falls	ID	1980	1	6.34	14	100	38
Philip	Philip	SD	1980	7	2.5	5.2	151	66
Pagosa Springs	Pagosa Springs	CO	1982	22	5.1	4.8	146	63
Idaho Capital Mall	Boise	ID	1982	1	3.3	18.7	150	66
Elko	Elko	NV	1982	18	3.8	6.5	176	80
Boise City	Boise	ID	1983	58	31.2	19.4	170	77
Warren Estates	Reno	NV	1983	60	1.1	2.3	204	96
San Bernardino	San Bernardino	CA	1984	77	12.8	22	128	53
City of Klamath Falls	Klamath Falls	OR	1984	20	4.7	10.3	210	99
Manzanita Estates	Reno	NV	1986	102	3.6	21.2	204	95
Elko County School District	Elko	NV	1986	4	4.3	4.6	190	88
Gila Hot Springs	Glenwood	NM	1987	15	0.3	0.9	140	60
Fort Boise Veteran's Hospital Boise	Boise	ID	1988	1	1.8	3.5	161	72
Kanaka Rapids Ranch	Buhl	ID	1989	42	1.1	2.4	98	37
In Search of Truth community	Canby	CA	2003	1	0.5	1.2	185	85
Bluffdale	Bluffdale	UT	2003	1	1.98	4.3	175	79
Lakeview	Lakeview	OR	2005	1	2.44	3.8	206	97

Solar heat

Use of solar heat for district heating has been increasing in Denmark and Germany^[51] in recent years.^[52] The systems usually include interseasonal thermal energy storage for a consistent heat output day to day and between summer and winter. Good examples are in Vojens^[53] at 50 MW, Dronninglund at 27 MW and Marstal at 13 MW in Denmark.^[54]^[55] These systems have been incrementally expanded to supply 10% to 40% of their villages' annual space heating needs. The solar-thermal panels are ground-mounted in fields.^[56] The heat storage is pit storage, borehole cluster and the traditional water tank. In Alberta, Canada the Drake Landing Solar Community has achieved a world record 97% annual solar fraction for heating needs, using solar-thermal panels on the garage roofs and thermal storage in a borehole cluster.^[57]^[58]

Low-temperature natural or waste heat

In Stockholm, the first heat pump was installed in 1977 to deliver district heating sourced from IBM servers. Today the installed capacity is about 660 MW heat, using treated sewage water, sea water, district cooling, data centers and grocery stores as heat sources.^[59]^[60] Another example is the Drammen Fjernvarme District Heating project in Norway which produces 14 MW from water at just 8 °C, industrial heat pumps are demonstrated heat sources for district heating networks. Among the ways that industrial heat pumps can be used are:

As the primary base load source where water from a low grade source of heat, e.g. a river, fjord, data center, power station outfall, sewage treatment works outfall (all typically between 0 ˚C and 25 ˚C), is boosted up to the network temperature of typically 60 ˚C to 90 ˚C using heat pumps. These devices, although consuming electricity, will transfer a heat output three to six times larger than the amount of electricity consumed. An example of a district system using a heat pump to source heat from raw sewage is in Oslo, Norway that has a heat output of 18 MW(thermal).^[61]
As a means of recovering heat from the cooling loop of a power plant to increase either the level of flue gas heat recovery (as the district heating plant return pipe is now cooled by the heat pump) or by cooling the closed steam loop and artificially lowering the condensing pressure and thereby increasing the electricity generation efficiency.
As a means of cooling flue gas scrubbing working fluid (typically water) from 60 ˚C post-injection to 20 ˚C pre-injection temperatures. Heat is recovered using a heat pump and can be sold and injected into the network side of the facility at a much higher temperature (e.g. about 80 ˚C).
Where the network has reached capacity, large individual load users can be decoupled from the hot feed pipe, say 80 ˚C and coupled to the return pipe, at e.g. 40 ˚C. By adding a heat pump locally to this user, the 40 ˚C pipe is cooled further (the heat being delivered into the heat pump evaporator). The output from the heat pump is then a dedicated loop for the user at 40 ˚C to 70 ˚C. Therefore, the overall network capacity has changed as the total temperature difference of the loop has varied from 80 to 40 ˚C to 80 ˚C–x (x being a value lower than 40 ˚C).

Concerns have existed about the use of hydrofluorocarbons as the working fluid (refrigerant) for large heat pumps. Whilst leakage is not usually measured, it is generally reported to be relatively low, such as 1% (compared to 25% for supermarket cooling systems). A 30-megawatt heatpump could therefore leak (annually) around 75 kg of R134a or other working fluid.^[62]

However, recent technical advances allow the use of natural heat pump refrigerants that have very low global warming potential (GWP). CO₂ refrigerant (R744, GWP=1) or ammonia (R717, GWP=0) also have the benefit, depending on operating conditions, of resulting in higher heat pump efficiency than conventional refrigerants. An example is a 14 MW(thermal) district heating network in Drammen, Norway, which is supplied by seawater-source heatpumps that use R717 refrigerant, and has been operating since 2011. 90 °C water is delivered to the district loop (and returns at 65 °C). Heat is extracted from seawater (from 60-foot (18 m) depth) that is 8 to 9 °C all year round, giving an average coefficient of performance (COP) of about 3.15. In the process the seawater is chilled to 4 °C; however, this resource is not used. In a district system where the chilled water could be used for air conditioning, the effective COP would be considerably higher.^[62]

In the future, industrial heat pumps will be further de-carbonised by using, on one side, excess renewable electrical energy (otherwise spilled due to meeting of grid demand) from wind, solar, etc. and, on the other side, by making more of renewable heat sources (lake and ocean heat, geothermal, etc.). Furthermore, higher efficiency can be expected through operation on the high voltage network.^[63]

Heat accumulators and storage

Increasingly large heat stores are being used with district heating networks to maximise efficiency and financial returns. This allows cogeneration units to be run at times of maximum electrical tariff, the electrical production having much higher rates of return than heat production, whilst storing the excess heat production. It also allows solar heat to be collected in summer and redistributed off season in very large but relatively low-cost in-ground insulated reservoirs or borehole systems. The expected heat loss at the 203,000m³ insulated pond in Vojens is about 8%.^[53]

With European countries such as Germany and Denmark moving to very high levels (80% and 100% respectively by 2050) of renewable energy for all energy uses there will be increasing periods of excess production of renewable electrical energy. Heat pumps can take advantage of this surplus of cheap electricity to store heat for later use.^[64]^[65] Such coupling of the electricity sector with the heating sector (Power-to-X) is regarded as a key factor for energy systems with high shares of renewable energy.^[66]

Heat distribution

After generation, the heat is distributed to the customer via a network of insulated pipes. District heating systems consist of feed and return lines. Usually the pipes are installed underground but there are also systems with overground pipes. The DH system's start-up and shut downs, as well as fluctuations on heat demand and ambient temperature, induce thermal and mechanical cycling on the pipes due to the thermal expansion. The axial expansion of the pipes is partially counteracted by frictional forces acting between the ground and the casing, with the shear stresses transferred through the PU foam bond. Therefore, the use of pre-insulated pipes has simplified the laying methods, employing cold laying instead of expansion facilities like compensators or U-bends, being so more cost effective.^[67] Pre-insulated pipes sandwich assembly composed of a steel heat service pipe, an insulating layer (polyurethane foam) and a polyethylene (PE) casing, which are bonded by the insulating material.^[68] While polyurethane has outstanding mechanical and thermal properties, the high toxicity of the diisocyanates required for its manufacturing has caused a restriction on their use.^[69] This has triggered research on alternative insulating foam fitting the application,^[70] which include polyethylene terephthalate (PET) ^[71] and polybutylene (PB-1).^[72]

Within the system heat storage units may be installed to even out peak load demands.

The common medium used for heat distribution is water or superheated water, but steam is also used. The advantage of steam is that in addition to heating purposes it can be used in industrial processes due to its higher temperature. The disadvantage of steam is a higher heat loss due to the high temperature. Also, the thermal efficiency of cogeneration plants is significantly lower if the cooling medium is high-temperature steam, reducing electric power generation. Heat transfer oils are generally not used for district heating, although they have higher heat capacities than water, as they are expensive and have environmental issues.

At customer level the heat network is usually connected to the central heating system of the dwellings via

[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

[55]

[56]

[57]

[58]

[59]

[60]

[61]

[62]

[63]

[64]

[65]

[66]

[67]

[68]

[69]

[70]

[71]

[72]