Residential Heating Techno-Economics – PV, Heat Pump and Solar Thermal vs District Heating

Problem and context

BRF Årstaterrassen relies entirely on Stockholm Exergi district heating for space heating and hot water. During summer (May–September), space heating demand drops to zero and the district heating system serves only domestic hot water — but the piping network still incurs 15–20% heat losses. Stockholm Exergi was contemplating whether the district heating connection could be deactivated for these five months and replaced by on-site solar technology.

The annual heating demand of the complex is 2,604.4 MWh/year (131 kWh/m² × 47 m² × 423 apartments). During the three peak summer months (June–August), heating is exclusively hot water: approximately 60 MWh/month per building. The current district heating cost for the five-month summer period is 52,898 USD (556,871 SEK) out of a 3.5 million SEK annual contract.

September was selected as the dimensioning month because it has the lowest solar irradiation of the five months. With 60% of the 3,460 m² rooftop available (2,076 m²), the daily solar energy resource in September is 1,145.5 kWh and the daily hot water demand is 2,000 kWh.

Scenario 1 — District heating (BAU)

The baseline scenario: Stockholm Exergi supplies hot water via the existing district heating network at an annual summer cost of 52,898 USD. LCoE is 26.4 USD/MWh — the lowest of all scenarios because infrastructure is already in place and no capital is required. Demand fulfilment is 100%. This is the cost benchmark all alternatives must beat to justify investment.

Scenario 2 — PV + electric boiler (PVEB)

The PV and electric boiler scenario covers the rooftop with 1,116 SunPower SPR-MAX7-445-PT panels (0.45 kW each, 1.86 m² each) and uses a 0.227 MW electric boiler to convert PV electricity to hot water. A battery bank provides power during non-solar hours.

The fundamental problem: an electric boiler has a COP of 1. Even with all 2,076 m² of roof covered in PV, September solar output (1,145.5 kWh/day) falls far short of the 2,000 kWh/day demand. The boiler wastes every unit of electricity as heat at a 1:1 ratio, leaving a daily deficit of 854.5 kWh that cannot be bridged by storage. This scenario was eliminated as infeasible.

Component costs: PV at $876/kW, boiler at $51.6/kW (with a $13,779 CAPEX from trendline extrapolation to 227 kW), battery storage at $518.5/kW. OPEX $6,899/year.

Scenario 3 — PV + heat pump (PVHP)

Replacing the electric boiler with an air-to-water heat pump (COP = 3) reduces the electricity required per unit of heat by two-thirds. The heat pump specification: 79.65 kW capacity driven by 75.66 kW peak electrical demand (0.227 MW peak heat demand ÷ COP 3). Total daily electricity demand drops from 2,000 to 666.7 kWh.

Air-to-water heat pump was chosen over ground-source because drilling in Stockholm city is impractical. The battery requirement falls to 364.6 kWh (nocturnal demand 249.4 kWh, adjusted for 90% battery efficiency, 95% inverter efficiency, 80% depth of discharge). With COP = 3 the PV array covers all summer demand using only 58% of the available roof.

Scenario 4 — Solar thermal collectors + TES (STCTES) — recommended

Solar thermal collectors convert sunlight directly to heat rather than via electricity, eliminating the electrical conversion loss entirely. The selected collector is the Wagner-Solar C20 AR-M glazed flat-plate unit: 85.4% optical efficiency, 96% light transmissivity, selective absorber coating, 2.61 m² per panel, operating at up to 232°C stagnation temperature.

System sizing: 309 collectors covering approximately 806 m² generate 5,148.5 kWh/day in September — more than 2.5× the 2,000 kWh/day demand. Thermal energy storage (TES) handles the nocturnal gap of 723.4 kWh, requiring 1,205.7 kWh storage capacity (after 75% TES efficiency and 80% depth of discharge), corresponding to a storage tank volume of 2.95 m³ (temperature swing from 25°C to 60°C).

The water-glycol heat transfer fluid runs through the collectors and passes heat to the TES during daytime surplus. During evening and night (18:00–06:00), stored heat covers the 723.4 kWh demand that cannot be met from collectors alone.

Winter contribution: applying the STC system’s capacity across the full annual daylight profile yields 188.9 MWh/year — 18.9% of the 720 MWh annual domestic hot water demand. This means the STC system is not merely a summer substitute but also offsets a meaningful share of year-round district heating.

Comparative summary

The district heating LCoE of $26.4/MWh is the lowest because there is no capital to amortise — the question is whether a 5-year capital investment can replace an ongoing OPEX. At $776/year OPEX versus $52,898/year for district heating, the STC system pays back in 5 years on OPEX savings alone. Beyond year 5, the cooperative saves approximately $52,000/year in summer district heating fees.

Sensitivity analysis

A ±20% CAPEX sensitivity sweep showed the PVEB scenario is most exposed to capital cost changes — its cost base is dominated by PV panels and a large battery bank. The DH scenario is most sensitive to OPEX changes, reflecting that its entire cost is operational with no amortisable capital. STC-TES remained lowest-cost across all CAPEX perturbations within the ±20% range.

Limitations

• September was modelled as the single worst-case design month; actual multi-month simulation would capture performance variation across May–September more precisely.
• Demand was assumed constant at the June–August baseline throughout summer; occupancy-driven peaks (e.g., mornings) were captured in the hourly profile but not inter-month variation.
• PV panel degradation, STC soiling losses and long-term TES insulation performance were not modelled.
• The air-to-water heat pump COP of 3 is a fixed summer-average assumption; actual COP varies with ambient temperature and hot water set-point.
• PVEB feasibility assessment is based on September; summer months with higher irradiation (June, July) would improve but not resolve its fundamental COP-1 limitation.