How can heat pumps be adapted to different building types and climatic conditions? This article presents a flexible air-to-water heat pump with integrated PCM storage, developed and validated within the LIFE ITS4ZEB project to support low-carbon heating solutions.
Diego Menegon, Eurac Research | LinkedIn profile
Matteo Campidelli, Innova
Silvia Ricciuti, Innova
Andrea Bernardi, Eurac Research
Fabio Bertoletti, Eurac Research
(Note: Opinions in the articles are of the authors only and do not necessarily reflect the opinion of the European Union)
The building sector accounts for a significant share of global final energy consumption and greenhouse gas emissions, particularly due to heating and cooling demands, which have driven increasing interest in high-efficiency heating, ventilation, and air conditioning (HVAC) technologies and thermal energy storage solutions. In the EU, around 40% of energy is used in buildings, and 80% of the energy used in EU homes is for heating, cooling, and hot water [1].
In this context, heat pumps combined with thermal energy storage represent a key strategy to improve energy efficiency, enable load shifting, and enhance the integration of renewable energy sources in both new and refurbished buildings. PCMs are particularly attractive for thermal energy storage in buildings due to their high latent heat density and ability to operate within narrow temperature ranges relevant for space heating, cooling, and domestic hot water production. When actively integrated with heat pump systems, PCM storage can improve system flexibility, reduce peak electricity demand, and support grid-responsive operation.
Within the EU LIFE project ITS4ZEB, modular and adaptable heat pump solutions integrated with PCM storage have been developed to address a wide range of building typologies, insulation levels, and climatic zones across Europe [2].
This article presents the laboratory evaluation of one ITS4ZEB solution, focusing on the performance of a modular air-to-water heat pump with integrated PCM storage under different control strategies to maximise self-consumption and self-sufficiency.
In the ITS4ZEB project, various solutions have been developed and demonstrated.
Figure 1. Solutions developed in ITS4ZEB.
The prototype selected for the test is a system built with four modules. The first module, installed at the top, is a ducted ATW heat pump with R290, while the other three modules are the PCM storage of a capacity of 8 kWh each and a transition temperature of 48 °C. The heat pump has a nominal capacity at A7/W45 of 2.4 kW with a maximum (booster) capacity of 3.0 kW.
The unit's control system is designed to maximise the self-consumption of locally generated photovoltaic (PV) electricity after accounting for the building's electrical consumption. A backup system will provide the remaining heat needed to cover the loads. Additionally, the unit can be set to be self-sufficient, minimising the backup consumption.
Figure 2 shows the modules delivered to Eurac Research. Figure 3 shows the assembled prototype, with the PCM modules connected to each other and to the heat pump using the quick-connect fittings provided by Innova. The complete system was installed in the calorimeter of the Heat Pumps Lab. The air ducts were connected to the climate chamber of the laboratory. The unit allows the air ducts to be connected either at the top or at the rear. For a practical installation, the air ducts were connected to the rear of the unit. The hydraulic circuits for space heating/space cooling (SH/SC) and DHW were connected using flexible pipes.
Figure 2. Delivery of the prototype to the Heat Pumps Lab at Eurac Research.
Figure 3. Assembly and installation of the prototype in the Heat Pumps Lab at Eurac Research.
Figure 4 shows the energy fluxes of the system. Depending on the control strategy configured in the heat pump, the system can charge the PCM storage and provide heat to the building using only electricity from the PV system or by also drawing electricity from the grid. When the system does not provide the necessary capacity, the backup system provides the remaining energy needed to cover the loads.
Figure 4: Functional scheme.
The KPIs are derived from the energy flows shown in this scheme. Self-consumption is defined as the portion of the energy consumed from local production, while self-sufficiency is defined as the portion of the load covered by the system. For this system, electrical self-consumption is defined as the portion of PV electricity consumed directly by the heat pump, and there is no need to define thermal self-consumption.
As self-sufficiency is defined as the portion of the load covered by internal production, in this system, it can be expressed in thermal, electrical, or total terms. Thermal self-sufficiency is defined as the portion of the thermal load covered by the heat pump. Electrical self-sufficiency is defined as the portion of the electrical consumption of the heat pump covered by the PV system. Total self-sufficiency is defined as the portion of the consumption (electrical and primary) covered by the PV production.
The unit was installed in the Heat Pumps Lab at Eurac Research [3] in accordance with EN 14511-3:2022 [4]. The test was performed by emulating the building load using a hardware-in-the-loop approach based on the method developed by Eurac [5,6].
The climate selected for the test was Rome. The thermal load has been defined with a simulation of a renovated multifamily building, and the system covered the loads of one apartment in the building. The average domestic hot water consumption is 3 kWh/day, corresponding to three tenants, while the heating load is 2540 kWh/year and the cooling load is 1440 kWh/year. The PV system is a 3 kWp plant that was simulated considering the consumption of typical electrical loads of the building.
To define the six-day test sequence, the 365 days were classified according to air temperature, load, and PV production. The classification is shown in Figure 5. With the k-medoid clustering [5], the six test days were selected from the annual file. The results are presented in Figure 6, which shows the time series of external air temperature, building load (positive values indicate heating demand, while negative values indicate cooling demand), and PV production. The table in Figure 6 presents the characteristics of the selected days in terms of average temperature, load, energy production, and number of represented days.
Figure 5. Annual distribution of load and PV production according to the daily average temperature.
Figure 6. Test sequence and characteristics of the selected days.
In a system with thermal storage, the state of charge of the storage typically depends on the season, but the test sequence reproduces consecutive days, which are normally temporally distant from each other during the year. Therefore, the presence of thermal storage requires a preconditioning phase when transitioning between a winter day and a spring or summer day. The preconditioning phase used the calendar day immediately preceding the selected day. For example, between days 81 and 174, the test considered day 173 as the preconditioning day for day 174.
The test sequence was repeated twice to test the two strategies. The first strategy considers the logic developed by Innova to use the heat pump to maximise the self-consumption of PV production and use the boiler as backup and minimise the electricity taken from the grid. The second strategy instead maximises the self-sufficiency of the heat pump and therefore minimises the gas consumption.
The test considered an emulated load for space heating: the return temperature and the thermostat call were defined from the simulation, while the DHW was defined with a fixed draw-off profile.
To better understand the results, the sequence was divided into the winter season and the summer season.
Figures 7 and 8 show the time series of energy delivered and electrical consumption for two winter days and one spring day. The left graphs represent the self-consumption strategy, while the right graphs represent the self-sufficiency strategy. The top graphs present the energy provided for the space heating (SH), the domestic hot water (DHW), and the heat produced by the unit (W). The bottom graphs represent the photovoltaic production (PV) and the power input of the unit (PI).
In winter, during the test with the self-consumption strategy, the unit was activated when the PV energy was available and started to charge the PCM storage. In this phase, the unit was limited to consuming the available PV energy. After the conclusion of PCM charging on day 1, the unit provided SH by discharging the storage. The low availability of PV allowed the system to cover only a small portion of the DHW and SH loads. On the second day, the storage was charged completely, and the SH load was covered during the evening.
During the test with the self-sufficiency, the unit operated continuously in heat pump mode, covering the load. The period during which the heat pump was unable to meet the space heating demand was attributable to the charging process of the PCM storage.
Figure 7. Time series of two winter days: self-consumption strategy (left) and self-sufficiency strategy (right).
During the spring test, when the self-consumption strategy was employed, the unit was activated when PV energy became available and fully charged the PCM storage. The storage was charged within three hours, with another charging cycle starting after six hours.
In the self-sufficiency test, the unit charged the storage every six hours, distributing the charging sessions throughout the day, including periods without available PV energy. Each charging cycle lasted less than one hour, as there were no restrictions on using only PV energy.
Figure 8. Time series of one spring day: self-consumption strategy (left) and self-sufficiency strategy (right).
Figure 9 presents the self-consumption and self-sufficiency KPIs distinguished in the winter and summer seasons. During the winter season, as the names of the strategies indicate, the self-consumption strategy maximised the self-consumption, and the self-sufficiency strategy maximised the thermal self-sufficiency. As the electrical self-sufficiency does not consider the contribution of the backup boiler, the self-consumption strategy achieves the highest electrical self-sufficiency because the heat pump does not try to use electricity from the grid.
In the summer test, the differences between the two strategies were reduced. Both strategies covered the thermal self-sufficiency, but the self-consumption strategy reached the highest value of self-consumption because it concentrated the charge during the hours with PV availability. The cooling load was concentrated in the period when the PV is available.
Figure 9. KPIs for the heating and cooling seasons under two control strategies.
Figure 10 shows the energy fluxes of the two strategies extrapolated from the test sequence to annual operation. The self-consumption strategy results in a very low consumption from the grid but requires the gas boiler backup, while the self-sufficiency strategy covers all the load with the heat pump system.
Figure 10. KPIs divided in the heating and cooling season for the two control strategies.
The paper presented the modular and flexible heat pump system developed in the ITS4ZEB project and the laboratory characterisation of a modular system composed of a ducted air-to-water heat pump with PCM storage modules. The test considered two control strategies, the first maximising self-consumption and the second maximising self-sufficiency.
The main differences between the two control strategies were observed during the winter season, while in the summer season, the strategies produced more similar results due to the simultaneous availability of PV energy and cooling demand. As the system can be fully thermally self-sufficient, the gas backup can be used only when it is economically convenient, considering a strategy that minimises the operating costs.
[1] EC, (2025). https://energy.ec.europa.eu/topics/energy-efficiency/energy-performance-buildings/energy-performance-buildings-directive_en.
[2] ITS4ZEB, (n.d.). https://its4zeb.eu/.
[3] Eurac Research, Heat Pumps Lab, (2020). https://www.eurac.edu/en/institutes-centers/institute-for-renewable-energy/pages/heat-pumps-lab.
[4] EN 14511-3:2022, Air conditioners, liquid chilling packages and heat pumps with electrically driven compressors for space heating and cooling. Part 3: Test methods, European Committee for Standardization, Brussels, Belgium, 2022.
[5] D. Menegon, A. Soppelsa, R. Fedrizzi, Clustering methodology for defining a short test sequence for whole system testing of solar and heat pump systems, in: SWC 2017, ISES, Abu Dhabi, UAE, 2017.
[6] D. Menegon, A. Soppelsa, R. Fedrizzi, Development of a new dynamic test procedure for the laboratory characterization of a whole heating and cooling system, Appl. Energy 205 (2017) 976–990. https://doi.org/10.1016/j.apenergy.2017.08.120.
