District heating, 5GDHC and UTES — Part 1 & Part 2

1. Executive summary

Heat and cold are energy. A thermal kilowatt-hour is as much a unit of energy as an electrical one: it can be stored, moved through a network, and delivered to a building, a greenhouse, or an industrial process. Heating and cooling account for roughly half of final energy demand in the European Union, and remain largely covered by gas, fuel oil and coal.

This compiled dossier (Part 1, Rev. 3-C) joins the 150-year history of urban district heating — from Roman hypocausts and Chaudes-Aigues (1334) through 1GDH–4GDH (Lund et al., 2014) to 5GDHC (Buffa et al., 2019) — with the 2026 state of the art in underground thermal energy storage (UTES: PTES, ATES, BTES, MTES, CTES, TTES, GeoTES). Meshed hydraulic topologies are a structural requirement for low-temperature bidirectional networks, especially on flat terrain.

Read together, capture (geothermal, PVT, waste heat, electricity surplus), store (UTES across days to seasons) and distribute (meshed 4GDH/5GDHC) are one system. Remove storage and a well-designed low-temperature network still buys fossil thermies in January. Remove the mesh and stored energy has no efficient path to use. Revision 3-C adds that 4GDH and 5GDHC are complementary architectures, not a sequential replacement, and a subsection on heat-pump refrigerant selection versus delivery temperature.

2. Historical origins (Part I §1.1)

Antecedents include Roman hypocausts and heated baths. A more direct precursor appeared at Chaudes-Aigues, France, around 1334, where geothermal hot water was distributed through wooden pipes to about thirty houses. Commercial modern district heating began in North America: Birdsill Holly’s steam system in Lockport, New York (1877) and the New York Steam Company in Manhattan (1882), still operating.

Europe followed later: Germany in the 1920s, the Soviet Union from the 1930s, China from the 1950s–1960s, Scandinavia from the early twentieth century (Vaasa 1902, Västerås 1920, Karlstad 1948). Oil crises of the 1970s accelerated water-based systems. Global heat delivery via district heating reached about 11.5 EJ in the mid-2010s, mostly China, Russia and the EU.

3. Generations 1GDH–4GDH (§1.2)

The Lund–Werner classification tracks carrier temperature, materials, efficiency and integration.

1GDH · ≈1880–1930

Steam at 120–200 °C in steel pipes, typically in ducts. High losses, safety risk, condensation at the consumer. Lockport, Manhattan; steam remains in parts of New York and Paris.

2GDH · ≈1930–1980

Pressurised hot water generally above 100 °C, CHP, site-insulated steel, large substations. Widespread in the Soviet Union, Eastern Europe and early Western systems.

3GDH · ≈1980–2020 (“Scandinavian”)

Water below 100 °C, often 70–90 °C supply. Factory-prefabricated pre-insulated pipes, compact substations, metering, biomass and waste. Most operating European networks.

4GDH · from ≈2020

Lund et al. (2014): typically 50–70 °C supply, efficient buildings, low-grade renewable and residual heat, smart thermal grids coupled to electricity, gas, storage and demand response. Hydraulic topology — radial, looped or meshed — becomes first-order.

4. Fifth-generation district heating and cooling (§1.3)

Buffa et al. (2019) define 5GDHC (cold district heating, anergy grids) as a water- or brine-based grid with hybrid substations and water-source heat pumps, operating so close to ground temperature that it is not suitable for direct heating. Buildings act as prosumers. Near-ambient loops (commonly 5–30 °C, sometimes ~40 °C) enable simultaneous heating and cooling on the same pipes, plastic pipes, and seasonal storage (ATES, BTES, pit).

Small ΔT versus ground means minimal distribution losses — sometimes gains. Hydraulic topology, especially meshing, is a first-order design parameter.

5. 4GDH and 5GDHC are parallel, not sequential (§1.3.1) — Rev. 3-C

Generational labels can be misread as if 5GDHC were simply mature 4GDH. Lund, Werner, Thorsen, Gudmundsson and colleagues (Energy 227, 120520, 2021) state the opposite: 5GDHC is a parallel, complementary development sharing 4GDH’s objectives, not a fifth stage that supersedes a fourth. For many existing high-temperature connected buildings, 4GDH remains more cost-effective and resilient.

An operator need not choose a single destination. A dense 60–70 °C network can keep a 4GDH backbone for older buildings and process heat, while new districts, greenhouses or data-centre clusters are served by meshed 5GDHC ambient loops through WSHP substations. Section 2.5.4 shows the same split in refrigerants: the best working fluid differs between the two temperature regimes.

6. Network topologies and meshing (§1.4)

Hydraulic architecture determines resilience, pressure-drop distribution, multi-source and prosumer integration, and the real performance of low- and ultra-low-temperature systems. High-temperature radial trees can work for 1GDH–3GDH. Low-ΔT bidirectional multi-source systems need meshing; without it, the theoretical gains of ambient operation are cancelled by bottlenecks and single-point failures.

• Resilience: a pipe or pump outage does not isolate whole districts.
• Hydraulic balancing across parallel branches; fewer cold spots.
• Prosumer and multi-source injection without creating bottlenecks.
• Lower average pressure drop and pumping energy despite higher mass flow at low ΔT.
• Local heat/cold exchange at 5–35 °C; shorter effective transport distances.
• Scalability: new branches and sources without redesigning the grid.

Meshing is therefore a structural condition for 4GDH/5GDHC at scale, especially on flat terrain — not an optional refinement.

7. Thermodynamic foundations in Part 1 (§1.5)

Part 1 states the governing relations in short form: heat transport Q̇ = ṁ·cₚ·ΔT; buried-pipe losses scaling with ΔT to ground; heat-pump COP rising when lift falls; exergy destruction falling at lower network temperature; Darcy–Weisbach pressure drop rising with velocity squared. Lowering temperature cuts losses and raises COP, but forces higher mass flow — which meshing offsets by parallel paths.

Derivations and worked numerical examples live in Part 2 (UC-DHC-THERMO-EN-2026). Readers who only need the headlines can stop at Part 1 §1.5; readers who want to reproduce the numbers should open the companion PDF.

8. Global status, benefits and challenges (§1.6)

Capacity remains concentrated in China, Russia/CIS and Europe. The move from 3GDH toward 4GDH and 5GDHC pilots is most advanced in Northern Europe, parts of Germany, Switzerland and the Netherlands (Mijnwater / Heerlen). Benefits: lower losses, higher residual and renewable shares, simultaneous heating/cooling, sector coupling. Challenges: building-side temperatures, pumping energy, prosumer rules, and — often under-recognised — insufficient meshing.

9. UTES — IEA Geothermal 2026 corpus and taxonomy (§2.1–2.3)

A September 2026 IEA Geothermal corpus (with GEOTHERMICA, IEA Energy Storage TCP, BGS and UK partners) treats system-level thermal storage as the flexibility that lets wind and solar be fully used. Heat stored in summer is winter energy; cold stored in winter is summer cooling. Part 1 covers pit (PTES), aquifer (ATES), borehole (BTES), mine-water (MTES), cavern (CTES), tank (TTES) and deep geological (GeoTES, Witter et al. 2026) storage.

Useful physics: stored energy scales with mass, heat capacity and ΔT. Efficiency and CAPEX ranges in the 2026 corpus show PTES as the industrialising European fleet (TREASURE), ATES/BTES as mature where geology allows, and flooded mines as 5GDHC-ready infrastructure reconversion.

10. European industrial evidence (§2.4)

PTES: the European pit fleet and the TREASURE industrialisation programme (Grant 101136095) are the scale-up path Task 45 describes. ATES and BTES are proven in the Netherlands, Sweden and beyond. Flooded mines (MTES) couple directly to 5GDHC, as at Heerlen. CTES and TTES fill niche high-insulation or tank roles.

11. PVT collectors, water-source heat pumps and refrigerants (§2.5) — Rev. 3-C

PVT hybrids produce electricity and heat from the same square metre. Coupled to a water-source heat pump they avoid a borehole field and a noisy outdoor unit — a realistic retrofit on dense stock. IEA SHC Task 60 is the reference; manufacturer SPF/SCOP claims (Triple Solar, DualSun) are labelled as such, distinct from independent measurement. Summer PVT heat peaks when buildings need it least: BTES, ATES or PTES convert that surplus into winter energy (Yeoju; USES4HEAT).

“Water-source” names the carrier loop, not the refrigerant. IEA HPT Annex 58 and Wu et al. (2020) set a delivery-temperature crossover: below about 100–120 °C, ammonia, propane and CO₂ outperform water as refrigerant; above it, water (R718) and R717/R718 cascades take over. A 5GDHC ambient loop at 35–55 °C and a 4GDH or process line at 90–150 °C should not share a single refrigerant family.

12. Buildings, horticulture, industry and Africa (§2.6–2.9)

Storage must be linked to 4G/5GDHC buildings and grids, not left as a stranded pit. Greenhouses are a ready offtake (Middenmeer, Maghreb). Industry can both dump waste heat into UTES and take process cooling. Africa is not a lesser echo of Europe: geothermal cascades and greenhouses, flooded mines (Witwatersrand, Gauteng, Free State), and post-harvest solar cold rooms (ColdHubs, Soko Fresh) already displace diesel. ARGeo C11 (Moroni, 19–24 October 2026) is the near-term venue.

13. Policy 2026–2028 and the three pillars (§2.10–3.2)

IEA ES TCP Task 45: design tools exist; interest surged with the gas crisis. Missing pieces are permitting, standards, bankability of stored energy, and treating LTES as flexibility infrastructure. Contacts include van Helden (AEE INTEC), Gauthier (PlanEnergi), Fournier (Newheat), Schmidt (Solites), Adl-Zarrabi (Chalmers). Directives cited: RED III (2023/2413), EED recast (2023/1791), EPBD recast (2024/1275), ETS2 from January 2028.

Three pillars: capture, store, distribute. Remove the middle pillar and the system burns gas on 15 January. Near-term roadmap: Pau/Newheat PTES, AFPG/BRGM screening of the Paris Basin and mining basins, a “seasonal GWh_th” line in Fonds Chaleur dossiers, ARGeo C11 contribution — then a French PTES or HT-ATES permit within 12 months, and a Franco-African UTES case at Enerstock 2027.

14. Conclusions (Part 1)

District heating has moved from late-nineteenth-century steam to low-exergy bidirectional 5GDHC. Ambient temperatures maximise COP and valorise discarded heat; meshed topologies are equally necessary if those gains are to be captured. UTES is how networks decouple from gas across the year. Remaining obstacles are permitting, bankability and regulatory accounting — not a missing technology.

Europe should industrialise PTES and ATES at TREASURE/Task 45 pace and connect them to meshed 4G/5GDHC. Africa should treat cascades, mines and cold chains as a distinct proven deposit. The unit of account is the thermal kilowatt-hour. Classify this physics as infrastructure, count it in RED III and the EED, and deploy it from the Paris Basin to the East African Rift.

15. Part 2 — purpose, scope and relation to Part 1

Part 1 already incorporates the historical, classificatory and topological content of the older standalone 5GDHC history dossier. This companion (UC-DHC-THERMO-EN-2026) keeps only what Part 1 does not cover in depth: derivations of the five governing relations in Part 1 §1.5, and worked numerical examples that quantify claims such as “losses often fall 75% or more” or “meshing lowers average pressure drop”.

Worked examples use explicitly stated illustrative assumptions (a consistent 1 MW / DN150 / 500 m case), not measured project data, and must not be cited as field measurements. Document revision inside the companion file is 2-C (typesetting of formulas); it is the companion to Part 1 Rev. 3-C (user file name Rev3c).

15.1 Heat transport and energy balance

Thermal power in a fluid stream is Q̇ = ṁ · cₚ · ΔT. For a given delivered power, lowering ΔT raises required mass flow (or pipe diameter). That is the central design trade-off of 4GDH/5GDHC. Part 2 derives the relation from the steady-flow energy balance and works a 1 MW example at several ΔT values.

15.2 Distribution heat losses

Steady radial conduction through cylindrical insulation gives loss per unit length proportional to (T_fluid − T_ground) and to 1 / ln((r+δ)/r). The practical levers are insulation thickness and conductivity — and, for operators, fluid temperature. Worked example: comparative distribution loss by generation under identical pipe geometry, reproducing the order of the “often 75% or more” reduction from 1GDH to 5GDHC.

15.3 Heat-pump COP — Carnot and real

Ideal lift COP_Carnot = T_hot / (T_hot − T_cold) (absolute temperatures). Real machines achieve a fraction of that (Carnot efficiency ratio). Smaller lift — ambient loop to 35–55 °C emitters — raises COP. The worked example uses the same 1 MW case and shows why 5GDHC substations and 4GDH/process HTHPs should not share one refrigerant family (see Part 1 §2.5.4).

15.4 Exergy of “low-grade” heat

Exergy is the work that could still be extracted versus the environment. Lowering network temperature reduces destroyed exergy per delivered kilowatt-hour even when energy (first-law) efficiency looks similar. The companion works a numerical comparison so “low-exergy grid” is a quantity, not a slogan.

16. Hydraulic losses, meshing, synthesis

Darcy–Weisbach: Δp ∝ f · (L/D) · ρ · v². Higher mass flow at low ΔT would explode pumping energy on a single path. Meshed parallel paths cut velocity per branch and therefore friction. The worked example contrasts a single DN150 path with a meshed pair under the same 1 MW / 500 m duty.

The five relations interact: lower T_f cuts losses, raises COP and cuts exergy destruction, but raises ṁ — which only meshing keeps pumpable. None of the formulas in isolation explains why 5GDHC requires meshing; the consistent numerical case does. All tabled figures in Part 2 are illustrative and reproducible, not field data.

16.1 Companion conclusion

Part 2 adds no history or advocacy — that is complete in Part 1. It adds traceability: every thermodynamic and hydraulic claim in Part 1 §1.5 and §1.4 now has a first-principles derivation and at least one worked example. Read together, Part 1 makes the technical and policy case; Part 2 supplies the arithmetic.

17. References and downloads

Base dossiers (never “official”). Keep the previous August 2026 history PDF. Full citation lists are Annex A of each PDF.

• Part 1: UC-DHC-UTES-EN-2026 Rev. 3-C (PDF) · DOCX.
• Part 2: Thermo-hydraulics companion (PDF) · DOCX.
• Previous: History dossier, Aug 2026 (PDF).
• Lund et al. 2014 (4GDH); Buffa et al. 2019 (5GDHC); Lund et al. 2021 Energy 227, 120520 (parallel generations); IEA HPT Annex 58; Wu et al. 2020; IEA ES TCP Task 45; TREASURE.

Present · DFKPS Base dossier · PDF Part 1 rev.3-C Companion · PDF Part 2 thermo-hydraulics E.04 Efficiency hub