Experimental Evaluation of Hybrid Renewable and Thermal Energy Storage Systems for a Net-Zero Energy Greenhouse: A Case Study of Yeoju-Si

Abstract

The implementation of renewable energy systems (RESs) in the agricultural sector has significant potential to mitigate the negative effects of fossil fuel-based products on the global climate, reduce operational costs, and enhance crop production. However, the intermittent nature of RESs poses a major challenge to realizing these benefits. To address this, thermal energy storage (TES) and hybrid heat pump (HHP) systems are integrated with RESs to balance the mismatch between thermal energy production and demand. In pursuit of clean energy solutions in the agricultural sector, a 3942 m² greenhouse in Yeoju-si, South Korea, is equipped with 231 solar thermal (ST) collectors, 117 photovoltaic thermal (PVT) collectors, four HHPs, two ground-source heat pumps (GSHPs), a 28,500 m³ borehole TES (BTES) unit, a 1040 m³ tank TES (TTES) unit, and three short-term TES units with capacities of 150 m³, 30 m³, and 30 m³. This study evaluates the long-term performance of the integrated hybrid renewable energy and thermal energy storage systems (HRETESSs) in meeting the greenhouse’s heating and cooling demands. Results indicate that the annual system performance efficiencies range from 25.3% to 68.5% for ST collectors and 31.9% to 72.2% for PVT collectors. The coefficient of performance (COP) during the heating season is 3.3 for GSHPs, 2.5 for HHPs using BTES as a source, and 3.6 for HHPs using TTES as a source. During the cooling season, the COP ranges from 5.3 to 5.7 for GSHPs and 1.84 to 2.83 for ASHPs. Notably, the HRETESS supplied 3.4% of its total heating energy directly from solar energy, 89.3% indirectly via heat pump utilization, and 7.3% is provided by auxiliary heating. This study provides valuable insights into the integration of HRETESSs to maximize greenhouse energy efficiency and supports the development of sustainable agricultural energy solutions, contributing to reduced greenhouse gas emissions and operational costs.

1. Introduction

The world is currently facing an unprecedented global food crisis. Since the 1950s, the global population has grown from 2.5 billion to 8 billion and is projected to exceed 9 billion by 2050 [1]. This exponential growth has compounded the worldwide problem of food production, making food security a global focus for sustainability [2]. Compared to the 2014 baseline, the percentage of the human population suffering from malnutrition increased by 10.6% and 10.8% in 2015 and 2018, respectively [3]. Hence, investments in agriculture have been increasing to reduce global hunger and ensure food security [4]. In South Korea, arable land for growing crops is scarce, as two-thirds of the country is comprised of hills and mountains [5]. This impediment to an open-field agricultural system coupled with the seasonal variation of outside weather conditions has invigorated the greenhouse industry [6]. Although greenhouses provide a steady environment for meeting global food demand, maintaining profitable business operation is severely hampered by the high costs of heating and cooling [7].

Primitively, the heating energy used in the agricultural sector is derived from fossil fuels, rendering the agricultural sector responsible for one-third of global greenhouse gas emissions [8]. For instance, 80–90% of the heating energy in South Korea comes from oil-based heaters [9] and, before joining the International Energy Agency (IEA), the country was ranked sixth in the world for oil consumption [10]. As a requirement from all IEA members to work toward global energy stability, the South Korean government introduced several policies, one of which is the “2030 Greenhouse Gas Roadmap” detailing a plan to reduce carbon dioxide (CO₂) emissions by 3.6 million tons from six major sectors, including the agricultural sector [11]. Moreover, during the United Nations (UN) conference of parties convened in Paris in 2015, the South Korean government committed to advancing renewable energy systems (RESs) as part of its pledge to contribute to the global effort to limit the average temperature increase to 1.5 °C by 2050 [12]. This initiative aims to achieve a 20% increase in RESs, develop energy-independent cities, and promote the widespread adoption of photovoltaic thermal (PVT) collectors [13].

The main obstacle to increasing RESs in most sectors is the difference between the demand for thermal energy and the availability of renewable energy sources that varies by season. For instance, the winter temperature in South Korea can be as low as $-$20 °C, and the summer temperature can be more than 30 °C [5], creating an imbalance that influences the energy consumption pattern in most buildings including greenhouses. To overcome the challenge of seasonal mismatch between thermal energy demand and supply in the heating and cooling energy, different thermal energy storage (TES) systems are gaining popularity [14,15,16]. In greenhouse applications, TES holds the potential for reducing loss and achieving significant energy savings [17] and the suitable choice is influenced by factors like storage capacity, duration, and the required temperature for energy supply and demand. While short-term TES (STES) like phase change materials and tank TES (TTES) are designed to meet the heating load for some periods, long-term TES (LTES) like borehole TES (BTES), pit TES (PTES), and aquifer TES (ATES) cover the imbalance between solar energy generation and heating energy requirements [18].

Extant studies on the implementation of hybrid renewable energy and thermal energy storage (HRETESS) differ depending on the available technologies, the needed energy levels, and the accessibility of renewable energy sources. Saloux et al. [19] investigated the sizing and control of a BTES and two TTES within a solar district heating system. The study revealed that by decreasing or increasing the TTES volume, energy savings up to 36% can be achieved. In China, a pilot ATES is designed and implemented for greenhouse heating, utilizing solar energy stored in the soil [20]. In comparison to other conventional solar heating systems, the integrated systems achieved energy savings of 27.8 kWh/m² annually when maintaining the greenhouse temperature above 12 °C throughout the year.

Irrevocably, the use of heat pumps in net-zero energy building (NZEB) application is steadily increasing, driven by the installation of various HRETESSs, which support the continuous operation of heat pump systems, alongside the abundant availability of thermal energy in the environment [21]. Some notable studies have implemented heat pumps with different HRETESS configurations and evaluate the system performance. Yildirim et al. [22] tested a hybrid system that included photovoltaic (PV) and ground-source heat pump (GSHP) for a net-zero energy greenhouse (NZEG) located in Turkey. The ratio of the PV electricity generated to the greenhouse’s energy demand (heating, cooling, and lighting) is very high and the electricity generated met the greenhouse energy demand by 33.2–67.2% in summer [22]. Similarly, Jradi et al. [23] assessed an underground soil-based thermal energy storage system (UTES) integrated with a PV-air source heat pump (PV-ASHP) system to supply heating needs to a 1000 m² residential floor in Denmark. The overall system heating coefficient of performance (COP_sys) is approximately 4.76, with energy efficiencies for a standalone PV system, combined PV-ASHP system, and PV-ASHP system incorporating a UTES to be 5.88%, 19.1%, and 22.2%, respectively.

Amidst all this, there is a growing adoption of hybrid heat pumps (HHPs) to address the limitations of single-source heat pumps. Paradoxically, ASHPs are designed to provide higher temperatures but have lower efficiency during the winter period as a result of the low outside temperature. An early study by Ito et al. [24] used outside air and water sources simultaneously to produce hot water. The dual heat sources achieve a higher heating coefficient of performance (COP_h) when both outside air and water sources are efficiently utilized until their combined temperature equals the evaporation temperature of the ASHP. Meanwhile, Kim et al. [25] explored the energy-saving capabilities of a hybrid GSHP utilizing a thermal effluent source for greenhouse heating in South Korea. Three greenhouse growth conditions with low, medium, and high temperatures are investigated, and their thermal performance and energy usage during the heating season are contrasted. The results suggest that the hybrid GSHP could realize 17–20% greater energy savings compared to the conventional GSHP system.

Undoubtedly, greenhouses, employed in controlled-environment agriculture, have the potential to diminish carbon footprints by incorporating RESs, yielding positive outcomes for both the environment and farmers [26]. The integration of TES with RESs enhances energy management efficiency by storing surplus energy during periods of excess capacity and releasing it during high demand. This reduces and optimizes overall energy usage. However, full-scale implementation of the integrated systems in the agricultural sector has been elusive. Moreover, studies on the performance of multi-TES systems with different renewable energy sources are currently lacking in the literature. Additionally, integrating HRETESSs in greenhouses offers a viable solution to lessen reliance on fossil fuels and improve energy efficiency given the pressing need for sustainable agricultural practices and the policy-driven push for the adoption of renewable energy across all sectors. Greenhouses can save a significant amount of energy while preserving ideal growing conditions by utilizing solar thermal (ST) and PVT collectors in conjunction with HHP and TES technology. The performance of such integrated systems in practice, especially in countries like South Korea that experience large seasonal fluctuations, is still a crucial research topic. By experimentally assessing a HRETESS installed in a NZEG in Yeoju-si, in the northwestern part of South Korea, this study helps bridge this knowledge gap. The results provide valuable insights regarding operational viability, system efficiency, and the possibility of expanding HRETESSs in other agricultural sectors to support energy independence and sustainable food production.

Thus, this paper presents the build-up and long-term performance test of a full scale HRETESS installed at Purme Social Farm (PSF) in Yeoju-si, South Korea, with emphasis on the TES performance, COP, and solar fraction. The HRETESS integrates solar and geothermal energy to meet the greenhouse thermal and electrical load. The generated electrical energy from the PVT is stored in battery packs, while the generated thermal energy from the PVT, ST, and heat pump systems is stored in a TTES, BTES, and night TES (NTES). Remarkably, the HRETESS reckoned on the strength of HHP systems by using the TTES and BTES as heat sources in the winter to provide heating energy and acting as an ASHP in the summer to provide cooling energy to the greenhouse. This study is distinguished by its real-world implementation and performance evaluation of a fully integrated, full-scale HRETESS within a commercial greenhouse environment, an aspect currently lacking in the literature, which predominantly depends on simulations or small-scale trials. The study examines the significant deficiency in long-term empirical data about solar thermal efficiency, seasonal COP fluctuations, and the function of thermal storage in facilitating net-zero energy objectives for agricultural facilities. The findings of the study offer useful insights for enhancing renewable energy adoption in the agriculture sector across diverse climatic conditions.

2. Materials and Methods

2.1. Purme Social Farm in Yeoju-Si

The designated experimental site is located in Yeoju-si, Gyeonggi-do, South Korea. Initially, PSF consisted of low-tech plastic greenhouses in various sizes for cultivating strawberries, watermelons, cucumbers, and pumpkins. Later, it underwent renovation, transforming into a self-sufficient multi-span glass greenhouse integrated with HRETESSs (Figure 1). The multi-span greenhouse, positioned in a north-south orientation, covers a total area of 3942 m² and is divided into Farm A and Farm B for the cultivation of cherry tomatoes. Farm A has dimensions of 32 m in width, 67.5 m in length, and 7.25 m in height, while Farm B measures 36 m in width, 49.5 m in length, and 7.25 m in height [27].

To optimize solar radiation absorption and minimize heat loss, the greenhouse utilized environmentally friendly covering materials [28]. Both the sides and roof of the greenhouse are clad with horticultural glass (4 mm thickness, 0.89 transmittance, and 0.08 reflectance) and Fluorine film (0.08 mm thickness, 0.92 transmittance, and 0.06 reflectance). In addition, the greenhouse incorporates two retractable horizontal screens, Tempa (0.31 mm thickness, 0.10 transmittance, and 0.65 reflectance) and Luxous (0.30 mm thickness, 0.58 transmittance, and 0.30 reflectance), along with a vertical screen, Obscura (0.34 mm thickness, 0.01 transmittance, and 0.64 reflectance), to enhance the greenhouse energy efficiency. Tempa is shut during daylight hours when solar radiation and greenhouse temperatures stay below 200 $\mathrm{W}{\mathrm{m}}^{-2}$ and 18 °C, respectively [29]. Simultaneously, Luxous is closed when both solar radiation and greenhouse temperatures fall below 150 $\mathrm{W}{\mathrm{m}}^{-2}$ and 15 °C, respectively. The thermal screen position, dimension of a single-span, and the sensor position in the greenhouse are shown in Figure 2, while

Table 1. Sensor specifications.

Parameter	Unit	Sensor	Sensor Precision
Outside temperature	$°\mathrm{C}$	HOBOMX1102A	$\pm 0.5 °\mathrm{C}$
Relative humidity	%	HOBOMX1102A	$\pm 0.5\mathrm{\%}$
Solar radiation	Wm−2	CMP3 pyranometer	$\pm 20$ Wm−2
Water temperature	$°\mathrm{C}$	I-sensor, PT 100	$\pm 0.3\mathrm{\%}$
Flow rate	LPM	KFCM-1000 K-2101083-2	$\pm 0.5\mathrm{\%}$
Wind speed	ms−2	Clima sensor, US	$\pm 5\mathrm{\%}$
Wind direction	degree	Clima sensor, US	$\pm 5\mathrm{\%}$
Outside pressure	hPa	PTB-220TS, VAISALA	$\pm 5$ hPa
Digital power meter	kW	Im-PRO II	$\pm 2\mathrm{\%}$

provides details regarding the characteristics of the installed sensors.

2.2. Overview of the Hybrid Renewable Energy and Thermal Storage Systems

The HRETESS installed at PSF integrates multiple renewable energy and heat storage technologies to improve greenhouse energy efficiency. The system includes 231 SNG-CS1 flat plate ST collectors with a total capacity of 139.2 kW, which harness solar energy for direct heating of the greenhouse and indirect utilization by HHPs through the TTES. Additionally, 117 Q-Peak L-G4.2 PVT collectors with a combined capacity of 43.8 kW generate both electrical and thermal energy. The electrical energy is stored in battery packs, while the thermal energy is transferred to the BTES through short-term thermal storage from photovoltaic thermal collectors (STTS-p). To address the intermittent nature of renewable energy sources, the HRETESS incorporates LTESs comprising TTESs and BTESs, as well as short-term energy storage (STES), which includes NTES, STES-p, and short-term thermal storage from solar thermal collectors (STES-s) of varying capacities [30]. For heating and cooling, the system utilizes two GSHPs with heating and cooling capacities of 76.9 kW and 80 kW, respectively. Moreover, four HHPs operating in both air-source and hydrothermal modes have 65 kW air heating, 56 kW air cooling, 74.2 kW hydrothermal heating, and 78.5 kW hydrothermal cooling capacities. Notably, two of the four HHPs use thermal energy from the TTES for heating operations, whereas the other two use thermal energy from the BTES for heating operations. The HHPs using TTES sources are jointly tagged as HP1, the HHPs using the BTES source are jointly tagged as HP2, and the two GSHPs are jointly tagged as HP3. HP1 and HP2 can automatically or manually switch to an air source, depending on the outside conditions and the source temperature. During such an operation, the HHP is labeled as an ASHP.

2.2.1. Description of the Thermal Energy Storage Systems

The TES at the experimental site includes LTES and STES. The LTES comprises the TTES and BTES, whereas the STESs include a 150 m³ NTES, a 30 m³ STTS-s, and a 30 m³ STTS-p. The NTES has two different thermal storages, unlike the STTS-s and STTS-p that store only hot water for actively charging the BTES. The NTES temporarily stores hot and cold water that is later discharged to the greenhouse via hot-water pipes and fan coil units during the heating and cooling period. The NTES is intermittently charged by the heat pump systems and discharged to the greenhouse daily to deliver hot and cold heat in order to maintain at least a 15 °C and 35 °C indoor temperature during the winter and summer season, respectively.

The TTES has a volume of 1040 m³ and is made of SS 275 low carbon steel, with two diffusers at the top and bottom made of STS 304 stainless steel as shown in Figure 3a. It is insulated with 100 mm and 200 mm urethane foam at the bottom and side. The stratification in the TTES is measured with nine resistance temperature detector sensors placed at a 1.5 m interval along the height of the TTES. The design of the TTES is such that it is used as a direct heat source for the greenhouse in early winter and serves as an indirect source for the HHP using a TTES source (HP1) during extremely cold periods.

The BTES utilizes approximately 28,500 m³ of the earth (Figure 3b) and is covered with a layer of insulation beneath the topsoil (Figure 3c). It includes 152 boreholes, 112 at the centre (cojoined with a green colour in Figure 3d) and 40 at the edge (cojoined with a blue colour in Figure 3d). Each of the boreholes are 30 m deep and spaced 2.5 m apart. The centre boreholes occupy 21,500 m³ of the BTES volume and are plumbed into four sections of seven parallel circuits, each with a string of four boreholes. The edge boreholes occupy 7000 m³ and are plumbed into four sections of two parallel circuits, each with a string of five boreholes. Each series string is designed in such a way that water flows from the boreholes closer to the centre, and to the boreholes closer to the edge during heat storage. However, water flows from the outer edge to the centre edge of each string during heat recovery. This configuration ensures high heat temperature storage at the centre of the BTES. Additionally, seven separate boreholes (depicted with a red colour in Figure 3d), three at the centre (tagged A, B, and C) and four at the edge (tagged D, E, F, and G), are used as an observation well to monitor the BTES temperature, while one borehole outside the BTES (10 m away and tagged H) is used as a reference well to monitor the ground temperature in comparison to the BTES temperature. Compared to the BTES that is 30 m deep, both observation and reference boreholes are at a 50 m depth and five PT1000 temperature sensors, separated 10 m vertically downward are placed in the boreholes.

2.2.2. Thermal Energy Flow and Operation of the HRETESS

Within the network of the integrated energy system, several temperature, power, flowrate, and pressure sensors are installed, and data collection (sampling rate at 30 s) and system operation are performed via an automatic control screen. The operation of the HRETESS is based on four modes in relation to the seasons in South Korea and each mode represents a distinct operation as depicted in Figure 4a–d. In the Figures, the red lines indicate the flow of the heated fluid, the blue lines represent the flow of the cooler fluid, and the black lines denote sections of the piping system where the thermal energy transfer is inactive or temporarily isolated. Figure 4a shows the initial TES operation in the TTES and BTES during spring (April–June). During this period, the ST collector uses a propylene glycol aqueous solution to absorb solar energy and store the absorbed heat in the TTES via a heat exchanger (HX01). The temperature of the TTES, which is within the range of 20–25 °C owing to the heating operation in the previous winter, gradually increases. Similarly, the PVT collector uses the propylene glycol aqueous solution as the heating medium to absorb solar energy and store the absorbed heat at the centre of the BTES via HX03 and STTS-p. During this period, natural ventilation is realized via the opening of the roof for the passive cooling of the greenhouse.

Figure 4b shows the active cooling operation of the greenhouse using the heat pump systems and the continuous TES in the TTES and BTES during summer (July–September). Owing to the continuous heat storage operation in the TTES, the maximum required temperature (70 °C) would have been reached, and the heat from the ST collector is channelled to the centre of the BTES via HX02 and STTS-s. When this occurs, the PVT collector that initially stores heat at the centre of the BTES via HX03 and STTS-p is directed to store heat at the edge of the BTES. The thermal energy from the ST and PVT collectors are initially accumulated in the STTS-s and STTS-p before being injected into the BTES. Conventionally, the ST collector produces higher thermal energy than the PVT collector. Therefore, high-temperature heat from the ST collector is stored at the centre of the BTES and low-temperature heat from the PVT collector is stored at the edges of the BTES. This is essential for reducing the heat loss from the BTES and maximizing the heating capacity [31]. Three sensors are placed in the STTS-s and STTS-p systems to create three stratification nodes and monitor the water tank temperature. During this period, the ASHP and GSHP are primarily used to cool the greenhouse. The solar heat stored in the TTES and BTES is not used as a heat source for the HHP during the cooling operation because HP1 and HP2 are being prepared for the heating operation. However, the thermal energy flow of the HHP is configured such that the TTES and BTES are used as heat sinks if the temperature of the cold water in the NTES produced by the GSHP and ASHP is not low enough to meet the cooling load. The inlet temperature at the source side of the HHP is controlled by a three-way valve. The valve returns the fluid to the HHP when the fluid temperature is high enough to still raise the temperature of the HHP’s evaporator and by passing the flow carrying heat from the TTES or BTES when the fluid temperature is not sufficient to raise the temperature of the evaporator. By raising the temperature of the HHP evaporator, the COP is increased compared with the ordinary heat pump [32]. Additionally, the NTES, intermittently charged and discharged with thermal energy daily serves a dual purpose of hot and cold store. The operation of the heat pumps through the heat pump provides cold water to the NTES and five sensors are fixed in the NTES to monitor the temperatures of the isothermal nodes.

Figure 4c shows how direct solar heat is supplied to the greenhouse during late autumn and early winter (October–November) from the TTES. Seemingly, the first heating stage of the greenhouse begins during this period, and the solar heat stored in the TTES through HX04 is directly supplied to the greenhouse.

Finally, Figure 4d shows how indirect solar heat is supplied to the greenhouse during winter (December–March) by using heat pumps. During this period, the functionality of the TTES for direct heating of the greenhouse would have been depleted because the internal temperature of the TTES would have dropped lower than the greenhouse supply setpoint temperature (50 °C). To provide heating energy to the greenhouse, HP1 and HP2 are used to provide hot water to the NTES. During the heating operation of the HHP, if the evaporative inlet source temperatures of HP1 and HP2 drop to <20 °C, the HHP would switch to the ASHP. Conversely, when the outside temperature is 0 °C and the relative humidity is 85%, the operation of the ASHP would stop and the GSHP would begin its operation. An average NTES water temperature of 50 °C is maintained during the heating period. In extreme weather conditions, when the daily heating load is higher than the threshold NTES can supply, an electric boiler installed between the NTES and the greenhouse increases the hot water input temperature to ~55 °C.

2.3. Performance Evaluation

The performance of the HRETESS is evaluated based on solar system efficiency, thermal storage efficiency, heat pump efficiency, and solar fraction as described below:

2.3.1. Solar System Efficiency

The ratio of the heat absorbed by the ST and PVT collectors to the amount of solar radiation absorbed by the collectors is defined by Equation (1), as provided by solar district guidelines [33]:

$${\eta }_{collector}={\eta }_0-{\alpha }_1\left[{\displaystyle \frac{{(T}_m-T_a)}{G}}\right]-{\alpha }_2\left[{\displaystyle \frac{{{(T}_m-T_a)}^2}{G}}\right],$$

(1)

${\eta }_{\mathrm{c}\mathrm{o}\mathrm{l}\mathrm{l}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r}}$ represents the solar collector efficiency, ${\eta }_0$ denotes the optical efficiency, ${\alpha }_1$ and ${\alpha }_2$ represent the first- and second-order coefficient heat losses, respectively, $T_m$ denotes the collector temperature ($°\mathrm{C})$, $T_{\mathrm{a}}$ represents the ambient temperature $(°\mathrm{C})$, and G is the solar radiation (${\mathrm{W}\mathrm{m}}^{-2})$. The parameters ${\eta }_0$, ${\alpha }_1$, and ${\alpha }_2$ are obtained from the standard performance test conducted by the Korea Laboratory Accreditation Scheme, testing number 203 [34] on a clear day at normal incidence. The coefficients (${\alpha }_1$ and ${\alpha }_2$) plot the collector efficiency with respect to the ratio of the temperature differential to the solar radiation and the difference between the outside and heating-medium temperatures.

2.3.2. Thermal Storage Efficiency

The LTES efficiency directly affects the energy system performance compared to the STES because the acquired solar energy is directly stored in the LTES and directly discharged to the greenhouse and indirectly used by the HHP. For instance, the TTES is charged via the ST collector, while it is discharged when solar heat is directly used for heating the greenhouse or indirectly when the TTES is used as a heat source for HP1. In contrast, the BTES is charged via the ST, PVT collector or both and only indirectly discharged when the stored heat is used as a heat source for HP2 during the heating season. Since the TTES directly supplies solar energy to the greenhouse and indirectly serves as a heat source for HP1, the efficiency defined in Equation (2) considers the total energy provided directly from the TTES to the greenhouse ($Q_{TTES,direct}$) and indirectly to HP1 ($Q_{BP1}$), and the internal energy change of TTES (${dQ}_{TTES}$) divided by the TTES useful energy gain ($Q_{TTES,input}$). In contrast, the efficiency defined in Equation (3) for the BTES considers the total energy the BTES provided indirectly to HP2 ($Q_{BP2}$) and the internal energy of the BTES (${dQ}_{BTES}$) divided by the BTES useful energy gain ($Q_{BTES,input}$). Meanwhile, a similar calculation for the BTES storage efficiency defined in Equation (4) is also used for all the STES systems since they are all also discharged via one output.

$${TTES}_{eff}={\displaystyle \frac{Total discharge}{Total charged}}={\displaystyle \frac{Q_{TTES,direct}+Q_{HP1}+{dQ}_{TTES}}{Q_{TTES,input}}}$$

(2)

$${BTES}_{eff}={\displaystyle \frac{Total discharge}{Total charged}}={\displaystyle \frac{Q_{HP2}+{dQ}_{BTES}}{Q_{BTES,input}}}$$

(3)

$${STES}_{eff}={\displaystyle \frac{Total discharge}{Total charged}}={\displaystyle \frac{Q_{output\left(i\right)}+{dQ}_{output\left(i\right)}}{Q_{input\left(i\right)}}}$$

(4)

where i stands for each of the STESs including, NTES, STTS-p, and STTS-s

2.3.3. Heat Pump Efficiency

Heat pump efficiency is mostly defined as a ratio of the energy delivered from the heat pump in terms of thermal energy in relation to the energy consumed by the heat pump in the form of electricity energy. The COP_h, cooling COP (COP_c), and source COP (COP_source), are calculated from Equations (5)–(7).

$${\mathrm{COP}}_{\mathrm{h}}={\displaystyle \frac{Q_h}{W}}$$

(5)

$${\mathrm{COP}}_{\mathrm{c}}={\displaystyle \frac{Q_c}{W}}$$

(6)

$${\mathrm{COP}}_{\mathrm{source}}={\displaystyle \frac{Q_{source}}{W_p}}$$

(7)

$Q_h$ denotes the rate of heat delivered by the heat pump (kWh) during the heating season, $Q_c$ denotes the rate of heat extracted by the heat pump (kWh) during the cooling season, $Q_{source}$ denotes the heat extracted or rejected at the source side of the heat pump (kWh) for HP1 and HP2, W is the heat pump power consumption (kWh), and $W_p$ is the total circulation pump power consumption operating within the energy systems to deliver thermal energy to the greenhouse (kWh).

2.3.4. Solar Fraction

The solar fraction is a measure of the proportion of the greenhouse energy demand that is met by solar energy. In this study, the direct solar fraction met by the TTES is defined by Equation (8) and the total solar fraction is defined by Equation (9):

$$\begin{gathered} {SF}_{direct}={\displaystyle \frac{Direct solar energy supplied to the greenhouse from TTES}{Total energy supplied to the greenhouse}} \\ ={\displaystyle \frac{Q_{TTES,direct}}{Q_{load}}} \end{gathered}$$

(8)

$$\begin{gathered} {SF}_{total}={\displaystyle \frac{Thermal energy supplied to the greenhouse from TTES, HP1, and HP2}{Total energy supplied to the greenhouse}} \\ ={\displaystyle \frac{Q_{TTES,direct}+Q_{HP1}+Q_{HP2}}{Q_{load}}} \end{gathered}$$

(9)

where $Q_{HP1}$ and $Q_{HP2}$ are the combined thermal energy supplied to the greenhouse by HP1 and HP2 in kWh.

3. Results and Discussion

PSF is designed to be an innovative solution to existing greenhouse structures that rely on fossil fuels in South Korea. The design is uniquely like the world-known Drake Landing Solar Community in Canada [31], with the exclusion of the STES,—which stores only cold water—and the inclusion of the TTES and PVT collector. Its construction was completed in June 2021, apart from the PVT collector, which started operation in June 2022. A summary of the integrated operation of the HRETESS is presented in

Table 2. Summary of the operation of the monitored system performance.

	1 October 2021–30 September 2022	1 October 2022–31 March 2023
Heating days	182	182
Cooling days	92	$-$
Horizontal solar radiation (MWh)	783	325
Incident solar radiation on ST (MWh)	571	227
ST collected solar energy (MWh)	280	101
PVT collected solar energy (MWh)	24	$-$
ST efficiency (%)	47	42
ST loop losses (%)	23	26
PVT efficiency (%)	46	–
PVT loop losses (%)	7.5	–
TTES efficiency (%)	36	92
BTES efficiency (%)	61
STTS-s efficiency (%)	90
STTS-p efficiency (%)	95
NTES efficiency (%)	92	98
Heat pump output (MWh)	738	581
Heat pump loss (%)	2	1
Solar energy supplied to greenhouse (MWh)	20.9	20.3
Solar fraction (%)	41	37
Boiler consumption (MWh)	28.3	59.5
Total electricity consumed (MWh)	205	170
Total energy consumed by greenhouse (MWh)	696	613
GH thermal losses (%)	3	5
System COP	3.6	3.5

, and the detailed results for each component of the HRETESS are explained in the following sub-section. The thermal losses presented in Table 2 are calculated as the difference between the thermal energy supply to each loop and the actual heat consumed by the loop.

The average COP_system during the first year decreases from 3.6 to 3.5. This reduction is attributed to the decrease in heating energy supplied to the greenhouse through the heat pump, from 543.4 MWh to 533.2 MWh, and the decrease in minimum outside temperatures from −18 °C to −20 °C. The decrease in outside temperatures also contributes to an increase in boiler consumption, rising from 28.3 to 59.5 MWh, and a decrease in solar energy supplied from the TTES, declining from 20.9 to 20.3 MWh, to meet the greenhouse heating energy demand. Consequently, there is a reduction in the solar fraction, decreasing from 41% in the first year to 37% in the second year. Hence, the investigated system directly produces 3.4% of the total heating energy supplied to the greenhouse from solar energy, 89.3% indirectly through electricity consumption from the heat pumps, and 7.3% from boiler consumption.

3.1. Solar Thermal Systems

During the period of investigation, the monthly solar energy incident on the ST collector ranged from 56.3 to 156.3 kWh/m² while that on the PVT collector ranged from 36.0 to 127.8 kWh/m². Based on the gross area of the ST and PVT collectors, the total incident solar energies on the ST and PVT collectors are 798 and 57 MWh, respectively, and the collected thermal energies by the ST and PVT collector loops are 381 and 24 MWh, respectively. As shown in Figure 5, the mean monthly efficiency of the ST collector ranged from 25.3% to 68.5%, with the peak occurring in July 2022 when the total acquired solar energy from the ST collector is 40.8 MWh. Similarly, for the period when the PVT collector is in operation, the mean monthly efficiency ranged from 31.9% to 72.2%, with the peak occurring in September 2022 when the total acquired solar energy is 6.1 MWh. In the years 2022 and 2023, the average efficiencies of ST collectors decreased from 47% to 42% and over the study period when the PVT was in operation, the average efficiency was 46%. The decrease in solar collector efficiency can be attributed to the difference in the study periods. While the first-year data span a complete growth season, including both winter and summer months, the second-year data were limited to only the winter months, resulting in reduced solar energy accumulation.

Continuous heat storage occurred in the TTES from the ST collector, and gradual heat storage occurred in the BTES from the ST and PVT collectors. During the solar energy storage period in the TTES and BTES, 64.8% of the thermal energy acquired from the ST collector is stored in the TTES, 11.6% is stored in the BTES, and 23.6% is lost in the ST collector loop pipe. Similarly, 92.5% of the solar energy acquired from the PVT collector is stored in the BTES and 7.5% is lost in the PVT collector loop pipe. The lower heat loss from the PVT collector is due to the lower temperature of the PVT loop compared with the ST loop. The TTES operates with a high degree of thermal stratification to allow the topmost node with the highest temperature and solar heat to be available for a direct heat supply to the greenhouse in early winter.

3.2. Thermal Energy Storage Systems

Figure 6 shows the mean monthly storage temperature of all TES systems, calculated as the average temperature of all sensors across the TES nodes, ranging from 10 °C to 90 °C. In October 2021, the average TTES temperature is 53 °C and it dropped to ~30 °C in November because of the direct heat supply to the greenhouse that occurred in the previous month. The TTES had a minimum temperature of 15 °C between January and March because of the indirect utilization of the water temperature by HP1 during the heating season. By April, the TTES temperature increased to ~20 °C due to solar absorption from the ST collector and continued to increase until the temperature peaked at ~70 °C. This high-temperature solar heat water is then supplied as a direct heat source at the beginning of winter. The temperature difference between the highest and lowest nodes in the TTES ranged from 40 °C to 70 °C and 10 °C to 60 °C, during the charging and discharging periods, respectively. The TTES and NTES temperatures exhibited a sinusoidal relation but with an opposite trend as shown in Figure 6. During the summer when the TTES temperature is at the peak of storing hot water, the NTES temperature is at the minimum of storing cold water with temperatures ranging from 15 °C to 17 °C, prepared by the ASHP and GSHP. Regardless of the season, the average BTES, STTS-s, and STTS-p temperatures range between 12 °C $\mathrm{a}\mathrm{n}\mathrm{d}$ 14 °C, 16 °C $\mathrm{a}\mathrm{n}\mathrm{d}$ 26 °C, and 15 °C $\mathrm{a}\mathrm{n}\mathrm{d}$ 30 °C, respectively.

The TES temperature is directly influenced by the charging and discharging of the storage systems. Figure 7 shows the total energy charge and discharge of the TES system. The variations in the LTTS differed from those of the STTS system. The NTES is designed to satisfy the daily greenhouse loads, and the total thermal energy charged within a given period is almost completely discharged during the same period. The thermal energies of the STTS-s and STTS-p systems are immediately discharged into the BTES when their setpoint temperatures are reached. The sudden fall in the TTES system thermal energy input in July and August is a result of the BTES charging via the STTS-s when the TTES target temperature is reached. Over the investigated months, 247.8 MWh and 60.5 MWh of energy is charged into the TTES and BTES, respectively, and 130.4- and 411.2-MWh energy discharged. The BTES exhibited a higher energy discharge than energy charge because it acted as a source for HP2 during the first year of operation when it is not charged. Naturally, the BTES requires an initial charging period to stabilize the ground temperature. From the total discharge energy from the TTES, 20.3 MWh of energy is directly supplied to the greenhouse in October 2021 and 2022. The total energies charged into the NTES, STTS-s, and STTS-p are 1297.7 MWh, 44.3 MWh, and 22.2 MWh, respectively, while the total energies discharged are 1232.4 MWh, 40 MWh, and 21 MWh, respectively.

The BTES is discharged in the first year when it is not charged by either the STTS-s or STTS-p. Therefore, accounting for the BTES efficiency for each year is difficult. However, during the charging and discharging period, the average efficiency of the BTES is 61.2%. Similarly, the STTS-s and STTS-p charging and discharging occur during the period of BTES charging, so STTS-s and STTS-p storage efficiencies for each year are also difficult to be accounted for. However, within the period when charging and discharging occurred, STTS-s and STTS-p average efficiency are 90% and 95%, respectively. Further, the average storage efficiency of the NTES and TTES over the considered period is 95% and 53%. The LTTS exhibits a lower efficiency than the STTS because the former must be heated to the highest temperature before heat is retrieved. The variation between the charging and discharging periods of the STTS and LTTS influences the storage efficiencies of these systems. The charging and discharging periods of the STTS are shorter, reducing the heat loss from the storage tank and increasing the TES efficiency.

3.3. Heat Pump Systems

The full-scale tests of the heat pump systems under different conditions across the season present a variety of results. In Figure 8, the average COP_h for HP1, HP2, and HP3 during the heating seasons are 3.6, 2.5, and 3.3, respectively, as illustrated. When comparing these COPs with their respective COP_source values, reductions of 0.4, 0.2, and 0.7 are observed, respectively. HP3 exhibits a higher COP_source reduction. This is primarily due to the fact that more work is required on the source side of the GSHP compared to the source sides of HP1 and HP2, which utilize TTES and BTES heat sources, respectively. The heat source from the TTES also contributed to a higher COP_h for HP1 compared to HP2 assisted with a borehole. This emphasizes the significance of solar energy in the application of heat pumps.

During the cooling season, the average monthly COP_c of the ASHP consistently increases from 1.83, 2.43, and 2.83 in July, August, and September, corresponding to a decrease in the mean monthly outside temperature from 26.9 °C, 25 °C, and 21.2 °C, respectively as depicted in Figure 9. In contrast, the GSHP records a higher mean monthly COP_c of 5.7, 5.4, and 5.3. This is attributed to the fact that the ground temperature is lower than the ambient temperature, resulting in a relatively higher COP_c for the GSHP compared to the ASHP. During the same period, the COP_source for the GSHP is 3.4, 3.5, and 3.7, respectively. For the three cooling months, the COP_source for the GSHP is 1.5 times higher than the average COP_source for the ASHP, primarily due to the higher power of the source pump in the GSHP. The source pump of the GSHP had a power of approximately 4000 kW, whereas the combined power of the two fans in the ASHP had a capacity of 1.8 kW.

The heat extraction rate differs from the heat rejection leading to a thermal imbalance in the ground. A ground thermal imbalance substantially impacts the soil temperature and causes the ground temperature to vary significantly over time, affecting the thermal performance of the GSHP. This accumulation of unbalanced heat mostly exceeds the thermal diffusivity of the soil. Hence, the soil temperature gradually increases. During the cooling season, there is an approximate 1 °C increase in ground temperature because of heat injection, as depicted in Figure 10. The ground temperature exhibited a sinusoidal pattern in conjunction with the outside temperature, reaching a minimum value of 15.3 °C before cooling when the outside temperature is 36 °C. However, during the cooling operation at a lower outside temperature of 32 °C, the ground temperature rose to 15.8 °C. This increase in ground temperature persisted as heat is consistently injected into the ground during the cooling operation of the GSHP.

3.4. Integrated System Performance: Long-Term Operational Insights

The efficiency of the long-term thermal energy storage systems—TTES and BTES—is significantly influenced by climatic conditions, as their charging and discharging dynamics are directly linked to solar energy availability and the greenhouse’s thermal energy demand, both of which are location-specific and crop-dependent. For instance, in this study, direct discharging of solar heat from the TTES began in October during nighttime, when the monthly average outdoor temperature fell below 13.6 °C. Had the ambient temperature dropped below 15 °C earlier, discharging from the TTES would have commenced sooner, consequently affecting its seasonal energy storage and retrieval efficiency.

In the case of the BTES, without active thermal charging in the first year, it functioned as a conventional ground-source reservoir: acting as a heat source in winter (being warmer than the greenhouse air) and a heat sink in summer (being cooler than the greenhouse air). However, starting in July 2022, deliberate thermal charging of the BTES using excess solar energy enabled it to store more thermal energy for use during colder months. This thermal pre-conditioning is essential for improving the responsiveness of the HRETESS, especially under sudden weather fluctuations when the TTES alone may not meet the greenhouse’s heating requirements. These findings demonstrate that the seasonal operational performance of both the TTES and BTES, and ultimately the overall system efficiency, is significantly affected by variations in climate—including outdoor temperature patterns, solar radiation levels, and seasonal heating/cooling demand.

The relatively small reduction in solar energy supplied from the TTES during colder periods—despite falling outdoor temperatures—can be attributed to several factors. First, the TTES temperature in the second year was maintained at a higher level than in the first year, resulting in greater thermal reserves available for discharge. Second, the greenhouse setpoint temperature was lowered during colder periods in the second year, reducing overall heating demand and helping preserve the contribution of solar heat, both directly from the TTES and indirectly via heat pumps. Additionally, the control system prioritizes solar-derived heat—whether directly or via heat pumps—before switching to auxiliary heating. The TTES’s high level of insulation further reduces thermal losses to the environment, even at low ambient temperatures, thereby preserving stored energy over extended periods. Nevertheless, this higher TTES temperature contributed to a decline in solar collector efficiency in the second year. As the storage temperature increased, the temperature difference between the ST collector surface and the ambient air widened, resulting in greater heat losses and reduced collector efficiency. However, this operational choice was necessary to ensure sufficient solar energy supply under colder outdoor conditions. Maintaining the same TTES temperature as in the first year would have led to significantly lower usable solar energy. Hence, a trade-off emerged between maximizing the quantity of stored solar energy for heating and minimizing losses from solar collectors. To mitigate this efficiency loss, strategies such as employing advanced selective coatings on solar collectors and implementing dynamic collector control based on real-time TTES temperature conditions could be beneficial.

The higher COP observed for the TTES compared to the BTES is primarily due to the TTES’s ability to maintain higher storage temperatures. This reduces the temperature lift required by the heat pumps, thereby improving efficiency. In contrast, the BTES operates at lower temperatures, requiring more electrical energy to raise the fluid temperature to the required level, which reduces the system COP. Additionally, the integration of HHPs with solar-assisted thermal energy storage greatly enhances the system’s ability to adapt to varying outdoor conditions. The HHPs can dynamically switch between air-source, solar-assisted, and ground-source modes depending on the availability of thermal energy. For example, although solar heat stored in the TTES and BTES is not the primary source during cooling operations, the system is configured to draw from these sources when the cold water supplied by GSHPs or ASHPs is insufficient. This configuration enables the system to maintain stable operation under a wide range of environmental conditions.

These findings have broader implications for climates with more extreme temperature variations. In colder climates, preheated long-term storage (such as BTES) can improve the COP of heat pumps, while in hotter regions, the adaptive use of HHPs—particularly their transition between ASHP and solar-assisted modes—can enhance cooling performance by leveraging seasonal or diurnal solar energy availability. Therefore, the operational principles demonstrated in this study can inform the design of flexible, climate-resilient energy systems for greenhouses and other temperature-sensitive applications worldwide.

To further optimize the system and reduce boiler consumption under decreasing ambient temperatures, several strategies can be implemented:

●

Lowering the greenhouse setpoint temperature by 1–2 °C during peak heating periods, without compromising crop health, to reduce overall energy demand.

●

Implementing predictive control algorithms to dynamically adjust heat pump operations based on real-time demand and weather forecasts, ensuring priority use of renewable sources.

●

Upgrading to solar-tracking ST and PVT collectors, rather than fixed systems, to increase solar energy capture, thereby improving TES charging and system performance during colder periods.

In addition, the incorporation of alternative renewable energy resources, such as biomass and wind energy, can enhance the resilience of the system. Expanding the long-term thermal energy storage capacity would allow for greater accumulation of surplus solar energy, further reducing reliance on auxiliary heating and increasing the solar fraction. Improvements in TTES insulation and optimization of charging strategies would also enhance long-term storage efficiency. Ultimately, sustaining a high contribution of solar energy is essential for meeting net-zero energy goals and ensuring the economic and environmental sustainability of greenhouse operations

4. Conclusions

The development and long-term performance of a hybrid renewable energy and thermal energy storage system (HRETESS) installed in a greenhouse in Gyeonggi-do, South Korea, are evaluated in this study. Despite challenging outdoor conditions at the experimental site, the energy burden previously placed on carbon-based fuels is successfully shifted to renewable energy sources through direct and indirect solar heat utilization combined with heat pump systems. Continuous data monitoring at the PSF is essential for system optimization and efficiency improvements throughout the operational period. The use of hybrid heat pumps (HHPs) proved effective in achieving a higher coefficient of performance (COP) compared to conventional heat pumps, demonstrating the advantages of integrating solar-assisted thermal energy storage. Notably, the TTES heat source contributed to a higher COP for HP1 compared to HP2, which is assisted by a borehole, underscoring the crucial role of solar energy in enhancing heat pump performance. The results of this show that the BTES operated as a low-temperature storage system between 12 °C and 14 °C, achieved an average storage efficiency of 61.2%, while the TTES which functioned as a high-temperature storage system between 15 °C and 70 °C, achieved an average storage efficiency of 52.6%. The efficiency of the ST collectors declined from 49% in the first year to 45% in the second year due to the increased operating temperature of the collector, affecting overall system performance. Additionally, the GSHP’s cooling COP is 1.7 times higher than its heating COP, attributed to the lower and more stable ground temperature during the cooling season, which facilitated efficient heat rejection and enhanced system performance. This study demonstrates the feasibility of integrating HRETESSs for greenhouse applications, reducing dependency on fossil fuels while optimizing energy efficiency. Future work will focus on further system modifications, long-term data analysis, and scalability assessments for other controlled environment agriculture.