Impact of Thermal Energy Storage on the Seasonal Performance of an Air-to-Water Heat Pump Under Real Microclimatic Conditions

Abstract

Air-to-water heat pumps (ASHPs) are a key technology for residential heating decarbonization; however, their seasonal performance is highly sensitive to outdoor temperature variability. Although thermal energy storage (TES) is widely recognized as a means of improving system efficiency, reported performance gains vary due to differences in climatic datasets, control strategies, and modeling assumptions. This study presents a systematic multi-year assessment of the impact of a water-based TES tank on the seasonal performance of a residential ASHP under measured microclimatic conditions. Hourly simulations were conducted for a single-family house at three locations in eastern Croatia using eight years (2018–2025) of measured meteorological data. Building characteristics, system configuration, and operating strategy were kept identical to isolate the influence of storage volume. TES integration reduced annual electricity consumption by 4.8–9.1%, with a multi-year average reduction of 7.02%, and consistently increased the seasonal coefficient of performance (SCOP) across all analyzed years and locations. The highest relative improvements occurred under less favorable microclimatic conditions, emphasizing the importance of diurnal temperature distribution rather than seasonal averages alone. A parametric analysis identified an optimal storage volume of approximately 1000–1500 L when both energy and economic indicators are considered. The results demonstrate that stable and reproducible seasonal efficiency gains can be achieved through a simple, non-predictive operating strategy under continental climatic variability.

1. Introduction

The global challenge of the 21st century is to ensure a reliable, affordable, and sustainable energy supply while minimizing environmental impacts. In this context, European energy policy aims to achieve climate neutrality by mid-century through substantial reductions in greenhouse gas emissions, which requires significant improvements in energy efficiency and increased deployment of renewable energy sources, particularly in the building sector [1,2,3]. Residential buildings play a central role in these efforts, as space heating and domestic hot water preparation account for a substantial share of final energy consumption.

Numerous studies have shown that building-level measures, such as optimized thermal insulation, can significantly reduce heating energy demand. However, building energy performance remains strongly affected by outdoor temperature variations. For example, recent research based on life-cycle cost analysis has demonstrated that appropriate selection of wall materials and insulation thickness can lead to considerable energy savings and reduced heating costs under heating-dominated climatic conditions [4].

Previous studies have shown that advanced thermal comfort models [5], incorporating both environmental and subjective parameters, can provide valuable input for the design and control of building energy systems, highlighting the importance of combining active thermal energy storage and heat pump operation with building envelope performance to ensure both energy efficiency and occupant comfort.

Air-to-water heat pumps (ASHPs) have been recognized as one of the most promising technologies for reducing greenhouse gas emissions in residential heating systems due to their ability to utilize renewable ambient energy and achieve high seasonal efficiencies [6]. However, the performance of ASHPs strongly depends on outdoor air temperature and operating conditions, particularly the supply temperature of the heating system. Under unfavorable ambient conditions, such as low outdoor temperatures, the coefficient of performance (COP) decreases significantly, resulting in increased electricity consumption and reduced seasonal efficiency [7,8,9].

Thermal energy storage (TES) can enhance the efficiency and flexibility of heat pump systems by temporally decoupling heat generation and demand, thereby reducing compressor cycling and enabling load shifting [10,11]. In residential applications, sensible TES (typically water-based) remains the most common solution [12,13], while latent TES can offer higher energy density and improved temperature stability but is still less widely adopted [14,15]. Although the performance benefits of TES–ASHP integration are well documented [16,17], reported improvements in seasonal indicators vary considerably due to differences in climate, storage volume, building characteristics, system configuration, and control strategy [18,19,20]. This variability complicates direct comparison between studies and highlights the need for systematic investigations that isolate the effect of TES under realistic operating conditions with constant boundary conditions.

Although numerous studies have investigated the integration of TES with ASHPs, a large share of existing research focuses on optimization-based or predictive control strategies. While such approaches can improve system performance, their practical implementation in typical residential buildings is often limited due to increased system complexity and data requirements. Consequently, the potential of simpler and more robust operating strategies under real climatic conditions remains insufficiently addressed in the literature [21,22,23].

Several studies have highlighted the importance of properly sized TES for mitigating partial-load losses in ASHP systems, particularly in high-efficiency residential buildings with low average thermal demand. Experimental investigations based on hardware-in-the-loop testing have shown that an appropriately designed thermal storage system can significantly reduce compressor cycling and improve seasonal performance, provided that storage thermal losses remain sufficiently low [24].

Moreover, many existing studies rely on Typical Meteorological Year (TMY) data or short simulation periods, which may not adequately capture the variability of real climatic conditions and their influence on seasonal ASHP performance. Since the efficiency of ASHPs is highly sensitive to outdoor air temperature, particularly during winter peak conditions, such simplifications can lead to an incomplete assessment of seasonal performance [25]. These limitations highlight the importance of analyses based on long-term measured climatic data that reflect realistic daily and seasonal temperature variations and enable a more robust evaluation of complementary solutions, such as TES, aimed at stabilizing system operation over the entire heating season [26].

The influence of total system water volume on the seasonal performance of ASHP systems has been widely investigated, yet the reported improvements and optimal storage volumes vary considerably across studies due to differences in climate, system configuration, and boundary conditions [27,28]. This further emphasizes the need for systematic studies that assess the impact of TES while maintaining constant building characteristics, system configuration, and control strategies [29].

To address these gaps, the present study investigates the influence of a water-based TES tank on the seasonal performance of a residential ASHP system using hourly simulations and eight years (2018–2025) of measured meteorological data. The analysis is conducted for three locations characterized by distinct microclimatic conditions while maintaining identical building characteristics, heating system configuration, and operating strategy.

Two system configurations are compared: a conventional ASHP system without TES and a configuration incorporating a buffer tank operated under a simple and practically implementable charging–discharging schedule that prioritizes operation during periods with higher outdoor air temperatures. The objective is to quantify the magnitude, stability, and reproducibility of seasonal performance improvements and to identify the storage volume range that minimizes annual electricity consumption while accounting for storage-related thermal losses.

Although the analysis is based on a representative single-family house, the applied methodological framework and the observed performance trends are transferable to similar residential ASHP systems operating under comparable continental climatic conditions.

The main contributions of this study are as follows:

(i)

A systematic eight-year assessment of TES influence using measured hourly microclimatic data rather than TMY datasets;

(ii)

Isolation of storage volume effects under identical boundary conditions across multiple locations;

(iii)

Quantification of seasonal efficiency gains achievable through a simple, non-predictive, and practically applicable operating strategy.

2. Materials and Methods

2.1. Heating System Description and Case Study Definition

This study considers a single-family house with a usable floor area of 170 m² as a representative residential case study for the analysis of ASHP-based heating systems under continental climatic conditions. The maximum heating capacity of the ASHP at an outdoor air temperature of −18 °C and a heating supply water temperature of 45 °C is 8.5 kW. In combination with an integrated electric auxiliary heater rated at 2.5 kW, the system fully covers the design heating load of the building. The design heat losses were calculated in accordance with the HRN EN 12831 standard [30] for a design outdoor temperature of −18 °C.

The system analysis was conducted under microclimatic conditions characteristic of the Slavonia region. For all settlements in Slavonia, the design outdoor temperature is −18 °C, except for the city of Požega, which has a lower design temperature of −20 °C due to its specific geographical location. Based on these design conditions, the selected heating system is considered representative for almost the entire Slavonia region.

Underfloor heating was adopted as the heat distribution system within the building, with a supply water temperature of 45 °C and a return water temperature of 35 °C. In this study, the heating season is assumed to last from 15 October to 15 April, during which the heating system is available for continuous operation.

The analyzed ASHP is a split-type system consisting of an outdoor unit containing the main refrigeration circuit components (compressor, evaporator, and expansion device) and an indoor hydraulic module incorporating the condenser (water-side heat exchanger) and circulation pump. The outdoor and indoor units are interconnected via refrigerant piping (gas and liquid lines), which enables flexible installation and significantly reduces the risk of hydraulic freezing under low outdoor temperature conditions, as no water circulates through the outdoor unit.

As illustrated in Figure 1, thermal energy is extracted from ambient air in the evaporator and upgraded through the vapor-compression cycle of the ASHP. The heated refrigerant transfers energy in the condenser to the water circuit located in the indoor hydraulic module. The produced thermal energy is subsequently supplied to the fully mixed TES tank, which is modeled as a uniform-temperature volume without thermal stratification.

The TES tank acts as a hydraulic buffer between the ASHP and the heating system, enabling temporal decoupling of heat generation and demand. From the storage tank, thermal energy is delivered to the underfloor heating system while the return flow is directed back to the TES tank.

The intrinsic water volume of the analyzed heating installation (including the underfloor heating circuits, connecting pipework, and the indoor hydraulic module) is 219 L, excluding the external TES tank. This value represents the baseline hydraulic capacity of the reference system without TES and was kept constant in all simulations. When TES is integrated, the total system water volume increases according to the selected storage tank capacity.

For reference, based on the commonly applied 30 min minimum compressor runtime criterion, the minimum equivalent water volume required for the analyzed ASHP capacity is approximately 0.37 m³ (≈366 L), assuming a usable temperature difference of 10 K (45–35 °C). The TES volumes investigated in this study (up to 1500 L, and extended to 5000 L in the economic assessment) therefore exceed the minimum operational requirement and primarily enable temporal load shifting under favorable outdoor temperature conditions.

The energy performance of the selected ASHP is characterized by its COP, which varies as a function of outdoor air temperature and heating supply water temperature. Figure 2 illustrates the variation in the COP of the ASHP as a function of outdoor air temperature and heating supply water temperature. Higher COP values are observed at higher outdoor air temperatures and lower supply water temperatures, reflecting more favorable operating conditions. The shaded regions indicate operating limits of the ASHP, where certain combinations of outdoor air temperature and supply water temperature are not permitted according to the manufacturer’s technical specifications.

The operating limits of the analyzed ASHP were defined based on manufacturer technical data. The maximum achievable heating supply water temperature was constrained as a function of outdoor air temperature, such that supply water temperatures of 55 °C and 60 °C were not permitted at outdoor air temperatures below −15 °C and −7 °C, respectively.

The ASHP was modeled to operate within its nominal performance range, with the COP defined solely as a function of outdoor air temperature and heating supply water temperature based on manufacturer technical data.

The thermodynamic refrigerant used in the analyzed unit is R410A, a widely adopted HFC refrigerant in residential ASHP applications. R410A operates at relatively high pressures and enables stable performance within the considered operating range, including supply water temperatures up to 45 °C under continental winter conditions. The manufacturer-provided performance data used in this study to define the COP as a function of outdoor air temperature and heating supply water temperature (Figure 2) are specific to this refrigerant and system configuration.

To assess the impact of TES, a hot water TES is integrated into the heating system. The TES is modeled to allow charging and discharging according to the predefined operating strategy described in Section 2.4.

2.2. Building Thermal Energy Demand Model

The thermal energy required to heat the single-family house over a given time period, i.e., to maintain the prescribed indoor air temperature, is calculated using Equation (1). A constant indoor air temperature was assumed throughout the heating season. The building thermal model was developed in accordance with HR EN ISO 13790 [31] and serves as a consistent reference framework for comparative system analysis.

A single-zone model was applied to quantify transmission heat losses, ventilation heat losses, and useful heat gains from internal sources and solar radiation. The model accounts for the thermal transmittance coefficients and surface areas of all relevant building elements, including external walls, glazed surfaces, doors, roof, and floor structures, as well as the air change rate of the heated space and a prescribed constant indoor temperature during the heating season. The entire heated space was represented as a single thermal zone assuming homogeneous heat distribution. The main building input parameters used in the thermal demand model are summarized in

Table 1. Main Building Input Parameters Used in the Thermal Demand Model.

Parameter	Value
Window U-value	1.00 W/m2K
Door U-value	1.05 W/m2K
Wall U-value	0.24 W/m2K
Floor U-value	0.26 W/m2K
Ceiling U-value	0.18 W/m2K
Thermal bridge correction coefficient	0.05 W/m2K
Internal heat gains	6 W/m2
Thermal zone assumption	Single thermal zone
Solar gains	Calculated using measured hourly solar radiation data
Ventilation losses	Calculated according to HR EN ISO 13790 [31] methodology based on total heated volume

.

Since hourly meteorological data are used in this study, the required thermal energy is calculated for a time interval of one hour, consistent with the temporal resolution of the climatic dataset.

$$Q=Q_{Tr}+Q_{Ve}-{\eta }_{gn}\left(Q_{int}+Q_{sol}\right)$$

(1)

The total thermal energy demand of the building, Q, consists of transmission heat losses through the building envelope, $Q_{Tr}$ and ventilation heat losses $Q_{Ve},$ reduced by effective internal and solar heat gains. Transmission heat losses depend on the thermal transmittance of each building component, its surface area, and the indoor–outdoor air temperature difference. In this way, heat losses through external walls, windows, doors, roof and floor are considered.

Ventilation heat losses are determined by the air change rate, the heated air volume, and the indoor–outdoor temperature difference. This component includes heat losses associated with the supply of fresh outdoor air.

Building heat gains consist of internal heat gains, $Q_{int}$, and solar heat gains, $Q_{sol},$ with their effective contribution adjusted by the heat gain utilization factor, ${\eta }_{gn}$. Internal gains include heat emitted by occupants, lighting, and electrical appliances. Solar gains depend primarily on incident solar radiation on external building elements, particularly glazed surfaces, as well as their orientation, surface area, and thermal properties.

Since the analyzed heating system consists of underfloor heating and the entire building is conditioned, distribution and emission subsystem losses are neglected. Therefore, the calculated thermal energy demand corresponds to the thermal energy delivered at the outlet of the ASHP during the considered time interval.

Based on the developed model and measured hourly meteorological data of outdoor air temperature and global solar radiation, the hourly heating energy demand of the building during the heating season was determined ($Q$ in Equation (1)). The calculation was performed for each individual hour over an eight-year period (2018–2025) for three analyzed locations: Gradište, Osijek, and Slavonski Brod, enabling a realistic assessment of the influence of microclimatic conditions on the building thermal load.

2.3. Heat Pump Performance Modeling

The relationship between the produced thermal energy $Q$ and the electrical energy $W$ consumed for heat pump operation is defined by the COP

$$COP={\displaystyle \frac{Q}{W}}$$

(2)

In this study, COP denotes instantaneous hourly values. The seasonal coefficient of performance (SCOP) is obtained by aggregating hourly delivered thermal energy and consumed electrical energy over the heating season.

The dependence of COP on outdoor air temperature and heating supply water temperature is defined based on the manufacturer’s technical data for the analyzed ASHP, as illustrated in Figure 2. The calculated hourly thermal loads were linked to the ASHP operating characteristics through this functional relationship. Consequently, the hourly electrical energy consumption W was determined and integrated over time to obtain the total annual electricity demand and SCOP for both the reference system without TES and the system with an integrated TES tank.

It should be noted that the COP representation is based on steady-state manufacturer data as a function of outdoor air temperature and supply water temperature. Dynamic effects such as part-load operation, compressor cycling, defrost processes, auxiliary electricity consumption, and control hysteresis were not explicitly modeled. However, both configurations were evaluated under identical modeling assumptions. Therefore, while absolute seasonal efficiency values may differ from real installations, the comparative performance assessment between configurations remains methodologically consistent.

2.4. Microclimatic Data and Analysis Methodology

Hourly meteorological data from three meteorological stations located in the Slavonia region—Gradište, Osijek, and Slavonski Brod—were used for the microclimatic analysis. These stations are situated in eastern Croatian counties and collectively cover a large portion of the Slavonia region. Meteorological data for the period from 2018 to 2025 were analyzed.

The microclimatic analysis was based on measured hourly outdoor air temperature data, which were used as input for the subsequent analysis of heat pump operation and TES strategies.

Using hourly meteorological data during the heating season, the temporal occurrence of daily maximum and minimum outdoor air temperatures was determined for each analyzed location. For each day within the analyzed period, the hour corresponding to the maximum and minimum outdoor air temperature was identified. The frequency of occurrence of these daily temperature extremes was evaluated for each hour of the day and for each month of the heating season.

Figure 3, Figure 4 and Figure 5 present the hourly distribution of the occurrence of daily maximum and minimum outdoor air temperatures during the heating season over the analyzed period for the Gradište, Osijek, and Slavonski Brod meteorological stations. These distributions were used to define representative daytime and nighttime temperature periods, which serve as input for the thermal energy storage operating strategy described in Section 2.4.

2.5. TES Modeling and Operating Strategy

To evaluate the influence of TES on heating system performance, the previously described system model was extended by incorporating a thermal storage tank and a predefined charging and discharging strategy. The storage tank was modeled as a fully mixed sensible TES with uniform temperature distribution. This assumption was adopted to ensure model simplicity and computational robustness, consistent with the study’s focus on practical applicability rather than detailed thermohydraulic optimization. Although thermal stratification may further enhance system performance, its consideration is beyond the scope of this work and does not affect the comparative assessment between systems with and without TES

Thermal losses from the storage tank were included in the model and assumed to be proportional to the temperature difference between the storage tank and the surrounding environment.

Thermal losses of the storage tank were calculated as:

Q_loss = U · A · (T_tank − T_amb),

(3)

where U represents the overall heat transfer coefficient of the tank insulation, A is the external surface area of the storage tank, and T_tank and T_amb denote the tank and ambient temperatures, respectively. The heat transfer coefficient UUU was assumed constant per unit surface area based on manufacturer insulation data, while the total heat loss coefficient UA varies with storage volume through geometric scaling of the cylindrical tank. Consequently, larger storage volumes exhibit lower surface-to-volume ratios, partially mitigating relative thermal losses.

For the TES-integrated system, a simple rule-based control strategy was implemented. During the charging phase, the ASHP operates at its maximum available heating capacity under given operating conditions until the storage tank temperature reaches 45 °C. Once this threshold is reached, the system switches to standard operation without additional storage charging. During the discharging phase, stored thermal energy is utilized until the tank temperature decreases to 35 °C, after which the ASHP resumes normal operation.

The operating strategy is not dynamically optimized; instead, its influence on seasonal performance is evaluated to ensure a simple, robust, and cost-effective control approach applicable to typical residential buildings. The numerical analysis was carried out in the MATLAB R2025b environment [32], using an hourly simulation framework.

By integrating hourly values, the total annual thermal energy required for space heating was determined for each analyzed year and location.

Storage volumes ranging from 50 L to 1500 L were considered, with a volume increment of 50 L. A wide range of charging and discharging time combinations was analyzed, assuming that both charging and discharging could occur at any hour of the day. For each combination of charging start time, discharging time, and storage volume, hourly electrical energy consumption was calculated for the entire heating season. The analysis was conducted for the full eight-year period and for all three locations. The simulation framework enables a consistent comparison of heating system performance with and without TES for different storage volumes and charging–discharging schedules.

The overall simulation workflow consisted of the following steps:

(1)

calculation of hourly building heating demand based on measured meteorological data;

(2)

determination of ASHP operating conditions and corresponding COP values;

(3)

TES charging/discharging logic implementation;

(4)

calculation of hourly electrical consumption;

(5)

seasonal aggregation of results;

(6)

repetition of the procedure for all storage volumes and operating schedules.

3. Results

This section presents the results obtained for the three analyzed locations—Gradište, Osijek, and Slavonski Brod—based on the eight-year dataset. Annual results are reported to enable a systematic comparison of the energy performance of the reference ASHP system without TES and the system incorporating a TES tank.

For each location and year,

Table 2. Results of the analysis of the impact of thermal energy storage on heating system energy performance for the period 2018–2025 at the Gradište, Osijek, and Slavonski Brod meteorological stations.

Year	Annual Thermal Energy Demand for Heating [kWh]	SCOP, Without a Storage Tank [-]	SCOP, with a Storage Tank [-]	Reduction in Electrical Energy Consumption [%]	Annual Thermal Energy Demand for Heating [kWh]	SCOP, Without a Storage Tank [-]	SCOP, with a Storage Tank [-]	Reduction in Electrical Energy Consumption [%]	Annual Thermal Energy Demand for Heating [kWh]	SCOP, Without a Storage Tank [-]	SCOP, with a Storage Tank [-]	Reduction in Electrical Energy Consumption [%]
	Gradište				Osijek				Slavonski Brod
2018	7433	3.05	3.19	4.79	7754	3.03	3.18	4.96	7855	3.02	3.20	5.88
2019	6299	3.18	3.38	6.32	6675	3.16	3.38	7.00	6870	3.12	3.37	8.28
2020	6860	3.19	3.39	6.22	7221	3.16	3.37	6.54	7402	3.12	3.36	7.70
2021	7688	3.13	3.34	6.77	7784	3.13	3.35	7.30	8059	3.08	3.32	8.01
2022	7349	3.13	3.36	7.29	7078	3.16	3.42	8.21	7018	3.14	3.42	9.11
2023	5746	3.26	3.49	6.99	6442	3.23	3.45	6.96	6432	3.19	3.47	8.75
2024	5137	3.27	3.51	7.23	6320	3.18	3.42	7.30	6268	3.17	3.44	8.36
2025	6238	3.20	3.40	6.14	7952	3.03	3.20	5.62	7346	3.10	3.31	6.80

first reports the annual space heating demand of the single-family house and the corresponding SCOP of the reference system. This reference performance level provides a consistent basis for evaluating the impact of TES integration.

The table also includes results for the TES-integrated system, evaluated under optimal operating conditions within the predefined and technically constrained range of storage volumes.

Within the analyzed range (50–1500 L), a storage volume of 1500 L resulted in the minimum annual electrical energy consumption in all considered cases. It is therefore identified as the optimal storage capacity within the investigated range. For clarity and conciseness, intermediate storage volumes are not explicitly reported in the table. Instead, Table 2 presents the SCOP values of both system configurations and the corresponding percentage reduction in electrical energy consumption achieved through TES integration for each location and year.

The results presented in Table 2 demonstrate that TES integration improves the seasonal performance of the ASHP system in all 24 evaluated cases, confirming the robustness of the identified efficiency gain under interannual climatic variability. Annual electricity savings ranged between approximately 105 and 195 kWh per dwelling, with an average value of about 145 kWh. Over the eight-year period, the relative reduction in electricity consumption varied between 4.8% and 9.1%, yielding a multi-year average reduction of 7.02% across all analyzed locations.

Although the baseline SCOP values of the reference system are relatively similar across the three locations, slightly lower values are generally observed in Slavonski Brod, reflecting less favorable microclimatic conditions. Following the integration of the 1500 L TES tank (identified as optimal within the investigated volume range), SCOP values increase systematically at all sites. The relative improvement is most pronounced in Slavonski Brod, indicating that the effectiveness of TES is strongly influenced by the frequency and magnitude of unfavorable outdoor temperature conditions rather than by seasonal average temperatures alone.

3.1. Statistical Summary of Seasonal Performance Improvements

To improve the clarity and comparability of the results, a statistical aggregation of the eight-year dataset was performed.

The mean reduction in annual electricity consumption amounts to: 6.47% (Gradište), 6.74% (Osijek) and 7.86% (Slavonski Brod).

The overall average reduction across all locations equals 7.02%.

The standard deviation of the relative electricity reduction across all 24 cases equals approximately 1.12 percentage points, indicating moderate interannual variability but stable performance improvement trends.

The coefficient of variation in the seasonal electricity reduction remains below 16%, further confirming that the observed efficiency gains are not driven by isolated climatic extremes but represent a systematic effect of thermal storage integration.

The corresponding mean SCOP increase is: +0.21 (Gradište), +0.21 (Osijek) and +0.25 (Slavonski Brod). In relative terms, this corresponds to an average SCOP improvement of approximately 7%, which is consistent with the observed reduction in annual electricity consumption and confirms the internal consistency of the simulation framework.

Table 3. Optimal charging and discharging times of thermal energy storage during the heating seasons (2018–2025).

Year	Gradište— Charging Start Time [h]	Gradište—Discharge Start Time [h]	Osijek—Charging Start Time [h]	Osijek— Discharge Start Time [h]	Slavonski Brod—Charging Start Time [h]	Slavonski Brod—Discharge Start Time [h]
2018	14	1	14	1	14	1
2019	15	24	14	24	14	1
2020	15	1	14	1	14	1
2021	14	1	14	1	14	2
2022	15	1	14	1	15	2
2023	14	1	14	1	14	1
2024	15	24	14	24	14	24
2025	15	1	14	1	14	2

presents the optimal charging start times and discharging start times of the TES tank during the heating season for the Gradište, Osijek, and Slavonski Brod meteorological stations over the period 2018–2025. The reported results correspond to the optimal system configurations that, for each individual year, yielded the minimum total electrical energy consumption within the predefined range of storage volumes and operating schedules.

The consistency of the optimal charging and discharging time instants presented in Table 3 further supports this interpretation. The predominance of afternoon charging (14:00–15:00) and late-night discharging (00:00–02:00) indicates that the optimal operating strategy is primarily governed by the diurnal distribution of outdoor air temperature.

This suggests that the observed efficiency gains arise from the temporal decoupling of heat generation and demand under systematically more favorable thermodynamic conditions, rather than from isolated interannual variations.

For all presented results, a TES tank of equal volume was applied, with a storage capacity of 1500 L identified as the optimal solution within the considered technically and spatially constrained storage volume range. The identified charging and discharging time windows correspond to the observed diurnal variation in outdoor air temperature during the heating season.

To further investigate the influence of TES tank volume on system performance, an additional analysis was conducted for a reference year and a representative location. This approach was adopted to avoid the influence of extreme climatic conditions and to ensure the general applicability of the conclusions.

The year 2020 was selected as the reference year within the analyzed period, as the annual space heating demand, the SCOP of the reference system, and the relative electrical energy savings were close to the multi-year average for all three locations. This selection minimizes the impact of exceptionally cold or unusually mild years and ensures methodological neutrality. Although 2020 was used as the reference year, the identified trends were additionally verified for colder and milder years, confirming that the qualitative relationships between storage tank volume, SCOP, and electrical energy savings remain consistent under interannual climatic variability.

Slavonski Brod was selected as the representative location because it consistently exhibited the least favorable microclimatic conditions and the highest relative benefits from TES integration, thereby allowing a clearer assessment of storage performance potential.

For the reference year 2020 and the Slavonski Brod location, a detailed analysis of the influence of storage tank volume on heating system performance was conducted. Storage volumes ranging from 50 L to 1500 L were evaluated with increments of 50 L. For each storage volume, the optimal charging start time and the optimal discharging start time were determined through an optimization procedure aimed at minimizing annual electrical energy consumption.

The results indicate that the optimal charging start time is independent of storage volume and consistently occurs at 14:00 for all analyzed cases. In contrast, the optimal discharging start time exhibits a clear dependence on storage capacity. For smaller storage volumes, discharge begins in the early morning hours (around 06:00), whereas increasing storage volume progressively shifts the optimal discharging time toward late-night hours, reaching approximately 01:00 for a storage volume of 1500 L.

As shown in Figure 6, the relative reduction in annual electrical energy consumption increases monotonically with increasing storage volume within the analyzed range. However, the rate of improvement gradually decreases at higher volumes, indicating diminishing marginal gains. The marginal reduction in annual electricity consumption decreases from approximately 0.54 percentage points per additional 100 L in the lower storage range (50–500 L) to less than 0.43 percentage points per 100 L above 1200 L. This clearly indicates energetic saturation and confirms that storage oversizing yields progressively smaller performance benefits.

The economic analysis was conducted using a simple payback time (SBTV) approach. Discounting effects, maintenance costs, and potential differences in storage tank lifetime were not included, as the objective was to compare relative economic trends between different storage volumes under identical boundary conditions rather than to provide a comprehensive investment appraisal.

To evaluate the economic feasibility of TES integration, the analysis was performed for the selected reference year and representative location, with the storage volume range extended up to 5000 L. For each storage volume, annual electrical energy savings were calculated based on the previously identified optimal operating strategy. Investment costs were estimated using manufacturer price data, including storage tank procurement, installation, and mounting costs.

The analysis was first carried out using the current reference electricity price in Croatia and subsequently extended to scenarios assuming electricity price increases of 25%, 50%, 75%, and 100%. It was assumed that the price increase applies exclusively to the energy tariff component, while all other components of the final electricity price remain unchanged.

The total investment cost comprises the insulated storage tank, an additional circulation pump, installation fittings, and mounting works. The cost of filling the system with water was considered negligible and included within the installation costs.

The individual investment components are:

−

1500 L insulated thermal storage tank: €935;

−

Additional circulation pump: €130;

−

Installation fittings: €70;

−

Installation and mounting works: €300.

For the 1500 L storage tank, the total investment cost amounts to €1435. The annual electricity consumption of the reference system without storage is 2372.3 kWh, while the system with the 1500 L storage tank consumes 2202.8 kWh, resulting in annual savings of 169.5 kWh. Assuming an electricity price of €0.2134/kWh, the corresponding annual financial savings amount to €36.18. Under these conditions, the resulting SPBT is approximately 39.7 years. The calculated SPBT indicates that, under current electricity prices and the applied cost assumptions, the economic return of TES integration is limited when evaluated solely through direct electricity savings.

However, this result is strongly influenced by the relatively low electricity tariff and the simple payback methodology applied. Since SPBT does not account for potential future increases in electricity prices, dynamic tariff structures, system lifetime considerations, or indirect benefits such as reduced compressor cycling and improved operational stability, the obtained value should be interpreted as a conservative estimate.

To assess the sensitivity of economic feasibility to electricity price variations, additional scenarios were analyzed assuming price increases of 25%, 50%, 75%, and 100%.

It was assumed that the electricity price increase applies exclusively to the energy tariff component, while all other components of the final electricity price remain unchanged. This approach enables an isolated assessment of the impact of electricity price variation on the economic performance of the system without introducing additional regulatory or market-related assumptions.

The results shown in Figure 7 indicate that the SPBT decreases markedly with increasing storage tank volume in the lower volume range, reaching a minimum at approximately 1000 L. However, SPBT for storage volumes of 1000 L and 1500 L are nearly identical, particularly under reference and moderately increased electricity price scenarios. This behavior indicates the presence of an economic saturation region, in which further increases in storage volume yield only marginal additional financial benefits relative to the increase in investment cost. When normalized per unit of installed storage capacity, the incremental financial return decreases substantially beyond 1000 L, confirming that energetic and economic optima converge within the same storage volume range (1000–1500 L). This convergence strengthens the practical relevance of the identified optimal sizing interval.

3.2. Key Quantitative Findings

The main quantitative findings of the multi-year analysis can be summarized as follows:

• TES integration reduces annual electricity consumption by 4.8–9.1%.
• The average multi-year reduction equals 7.02%.
• SCOP increases by approximately 0.22 (≈7%).
• The highest relative improvements occur under less favorable microclimatic conditions.
• Energetic saturation is observed above approximately 1000–1500 L storage volume.

4. Discussion

The observed improvements in seasonal performance indicators are closely related to the microclimatic characteristics of the analyzed locations and are consistent with findings reported in previous studies on heat pump systems coupled with TES. Several authors have shown that the effectiveness of TES strongly depends on the temporal distribution of outdoor air temperatures and the resulting operating conditions of the ASHP rather than solely on seasonal average temperature levels [19,23,25].

Locations characterized by frequent low nighttime temperatures, such as Slavonski Brod, exhibit lower baseline SCOP values in the reference configuration, confirming the strong sensitivity of ASHP performance to ambient temperature variations [23]. At the same time, these locations achieve greater relative energy savings when TES is integrated, indicating that the benefit of storage becomes more pronounced under less favorable microclimatic conditions. This observation is consistent with the findings of Arteconi et al. [19] and Wu et al. [25], who reported higher relative performance gains from TES in climates characterized by larger daily temperature fluctuations.

The identified charging and discharging windows provide clear thermodynamic justification. Afternoon charging shifts ASHP operation toward periods of higher evaporator temperature and reduced compression ratio, thereby increasing instantaneous COP. Nighttime discharge reduces compressor operation during periods associated with low ambient temperatures and reduced efficiency. This temporal decoupling between heat generation and demand represents the primary mechanism through which TES enhances seasonal performance, as emphasized in previous review studies [11,17].

The dependence of discharge timing on storage volume reflects physical storage capacity constraints. Smaller TES volumes require earlier discharge due to limited thermal reserve, whereas larger tanks sustain heat delivery over extended nighttime periods, enabling more effective avoidance of ASHP operation under low-COP conditions. Although increasing TES volume improves seasonal efficiency, the marginal benefit decreases progressively.

The economic assessment further extends the analysis by evaluating SPBT for storage tank volumes up to 5000 L. Although volumes up to 5000 L exceed typical residential constraints, their inclusion allows clearer identification of energetic and economic saturation effects. SPBT decreases markedly with increasing storage volume in the lower range, reaching a minimum at approximately 1000 L. However, SPBT values for 1000 L and 1500 L are nearly identical, particularly under reference and moderately increased electricity price scenarios.

Beyond 1500 L, further increases in storage volume do not lead to proportional reductions in SPBT despite additional electricity savings. This behavior indicates the presence of an economic saturation region in which incremental investment costs outweigh the marginal financial benefits.

Although the results clearly demonstrate the positive impact of TES on the seasonal performance of ASHP systems, several limitations should be acknowledged.

First, the applied TES operating strategy is intentionally simple and does not rely on predictive or optimization-based control algorithms. While advanced strategies such as model predictive control may achieve additional efficiency gains, they require increased system complexity, higher data availability, and greater control effort. The adopted approach therefore prioritizes robustness and practical applicability in residential installations. Although this may limit the maximum achievable performance improvement, it enhances real-world relevance and reproducibility.

Second, the analysis assumes a fixed electricity price structure and does not consider time-varying tariffs or demand-response incentives. The inclusion of dynamic electricity pricing could further improve the economic attractiveness of TES by enabling more effective load shifting. However, this aspect was intentionally excluded in order to isolate the influence of microclimatic conditions and storage volume on system performance.

Third, the results are based on a single representative building typology equipped with underfloor heating. Although absolute energy savings and optimal storage volumes may vary for buildings with different thermal characteristics or heat emission systems, the identified trends and relative performance improvements are expected to remain valid for residential buildings operating under similar continental climatic conditions.

Finally, the study focuses on locations within the Slavonia region, characterized by a continental climate with pronounced diurnal temperature variations during the heating season. Quantitative extrapolation to regions with substantially different climatic conditions should therefore be approached with caution. Nevertheless, the applied methodological framework and the identified relationships between microclimatic conditions, storage operation timing, and seasonal ASHP performance are transferable to regions with comparable climate characteristics.

Although buffer tanks are often recommended primarily to reduce compressor cycling, the storage volumes evaluated in this study, 1000–1500 L, exceed minimum hydraulic requirements. In the present context, their primary function is seasonal performance enhancement through temperature-driven load shifting, as consistently demonstrated across eight years of measured climatic data.

Limitations of the Study

The present study is subject to several methodological limitations.

First, ASHP performance was modeled using manufacturer steady-state COP data as a function of outdoor air temperature and supply water temperature. Dynamic effects such as part-load degradation, compressor cycling, defrost processes, auxiliary electricity consumption, and control hysteresis were not explicitly modeled. While this simplification may influence absolute SCOP values, both the reference and TES-integrated systems were evaluated under identical assumptions, ensuring methodological consistency in relative performance comparison.

Second, TES was modeled as a fully mixed volume without thermal stratification. Although stratification may influence real-system performance, the adopted approach ensures computational robustness and transparent isolation of storage volume effects.

Third, the economic evaluation was limited to SPBT. Discounting effects, net present value, and dynamic tariff structures were not included, as the primary objective was comparative energy-performance assessment rather than full life-cycle economic optimization.

Finally, the study considers a representative single-family house operating under continental climatic conditions. While the identified performance trends are expected to remain valid for similar residential buildings, quantitative extrapolation to other building typologies or climate zones should be undertaken with caution.

5. Conclusions

This study evaluated the impact of TES integration into a residential ASHP system using eight consecutive years of measured hourly microclimatic data. By maintaining identical building characteristics, system configuration, and operating strategy across three locations, the applied framework enabled a consistent and unbiased assessment of storage volume influence.

The results demonstrate that TES systematically improves seasonal performance under continental climatic conditions. Annual electricity consumption was reduced by 4.8–9.1%, with a multi-year average reduction of 7.02%, corresponding to a mean SCOP increase of approximately 0.22 (≈7%). Performance improvements were observed in all analyzed years and locations, confirming the stability and reproducibility of the storage-induced efficiency gain.

The effectiveness of TES was shown to depend primarily on the diurnal distribution of outdoor air temperature rather than seasonal average values. A simple, non-predictive charging–discharging strategy was sufficient to shift ASHP operation toward thermodynamically favorable periods, thereby enhancing seasonal efficiency without increasing control complexity.

When energetic and economic criteria were jointly considered, an optimal TES volume range of approximately 1000–1500 L was identified. Beyond this range, additional storage capacity resulted in diminishing marginal energetic and economic benefits, indicating the presence of an energetic–economic saturation region.

Overall, the presented multi-year analysis provides robust quantitative evidence that appropriately sized thermal storage can deliver stable and practically achievable seasonal efficiency improvements in residential ASHP systems under real microclimatic variability, even without advanced predictive control strategies.

Future research should extend the present framework toward dynamic electricity pricing models and advanced dynamic performance modeling in order to further assess system flexibility and long-term techno-economic performance under evolving energy market conditions.