Techno-economic assessment of a novel hybrid system of solar thermal and photovoltaic driven sand storage for sustainable industrial steam production

19 de julio de 2023

Highlights

•
The study presents a novel system combining solar thermal collector, pressurised water storage and PV driven sand storage for steam generation in food & beverage industry.
•
The optimal combination of parabolic trough collector and PV storage systems results in the lowest cost of heating and significant land area savings compared to if the technologies are used individually.
•
The hybrid system results in 16 % decrease in cost of heating compared to boiler-based system.
•
Technology collaboration is key to achieve high renewable heating fraction in industries and represent significant opportunities.

Abstract

Decarbonising industrial heat is a significant challenge due to various factors such as the slow transition to renewable technologies and insufficient awareness of their availability. The effectiveness of commercially available renewable heating systems is not well defined in terms of techno-economic boundaries. This study presents a techno-economic assessment of a novel system designed for steam production at a food and beverage plant. The proposed system is combines parabolic trough collectors with pressurized water thermal storage and photovoltaic-driven high-temperature sand storage. The technological components within the hybrid system complements each other both economically and practically, resulting in cost and land area savings. To evaluate the proposed system, simulations were performed using a model developed in TRNSYS and Python. The combined system exhibits better economic and land use performance than when these technologies are used individually. Specifically, the system has a high solar fraction of 90% while remaining competitive with the existing boiler fuel cost. The study emphasizes the importance of multi-technology approaches in developing practical solutions for industrial heat decarbonization. The findings can guide industries in a transition to sustainable heat sources and contribute to global efforts in mitigating climate change.

Keywords

Parabolic trough collector

Photovoltaic system

Sand storage

Water storage

Economic analysis

1. Introduction

Heat is constituting nearly 50% of the global final energy consumption, significantly higher when compared to energy consumption in other sectors such as transportation (30 %) and electricity (20 %) [1]. Heat is consumed in many sectors such as industries, residential and commercial buildings. Industrial heating is responsible for nearly 51 % of the total consumption, increasing greenhouse gas emissions [2]. In the year 2021, Global CO₂ emissions from fuel combustion in industrial processes reached their peak annual value of 36.3 gigatons. Industries generally rely on internal energy production, typically on the steam boiler systems for their heat production. In larger industries, it is common to have internal electricity production as well. The most used primary energy source is fossil fuels. The growing awareness of the detrimental effects of fossil fuels has for decades pointed towards the necessity to bring forth alternate, non-exhaustible energy sources that can reduce CO₂ emissions.

Recently, the industrial fuel supply has been particularly stressed due to the geopolitical situation. Especially in the EU, the Ukraine war exposed industries to the tide of increasing fuel prices. Natural gas prices reached a record level in 2021, and the trend of increasing prices is worrisome for the future [3]. The average natural gas price in Europe is expected to be higher than in last decade to shortage of the supply [4]. Similar increasing fuel price trends can be seen in South Africa, where the coal price has reached a record level of 330 USD/ton, the highest ever in the last 40 years [5]. Industries face challenges in securing high-quality coal for their boilers, as the preference for such coal is to export it to Europe. In Northern Europe, wood chip prices have doubled in the past years, reaching a value of 50–70 €/MWh, thus increasing industrial heating prices [6].

The use of renewable heating technologies can address the current geopolitical challenges in the fuel supply and provide sustainable solutions for industries while promoting energy security. Currently, renewable heat accounts for only 11% of the overall heating supply [7]. In contrast, the share of renewables in electricity generation is 28.3%, and for transport, it is 3.7%. To address the issue of climate change, the European Union has proposed a legal framework called “Fit for 55,” aiming to reduce net greenhouse gas emissions by 55% by 2030, when compared to GHG levels in 1990 [8]. This proposal focuses on energy-efficient buildings and industries, with various initiatives to make the EU more environmentally friendly. The renewable energy directive is initiative within “Fit for 55” package that aims to increase renewable heating and cooling by 1.1% per year. To achieve this goal, heat pumps (HP) and solar-driven heating systems using photovoltaics or solar thermal are considered key technologies to lead the way towards a decarbonized energy system.

Nearly 30% of the total process heating in EU industries requires temperature below 150 °C [9]. The industrial sector relevant to this temperature range includes brewery, food & beverage, textile, pharmaceutical, dairy, etc [10]. Several legislations aims to ban the use of fossil fuel boilers for process heating up to 200 °C and replace them with other energy-efficient systems [11]. To achieve this aim, there are several technologies available in the market that can provide this temperature range with low carbon emissions, such as solar thermal (ST) collectors, renewable electricity, such as PV/PVT, high-temperature heat pumps (HTHP), and boilers utilizing green fuels like waste biomass or biogas [12], [13].

Many multinational companies are now looking to solutions for renewable heating. Specifically, industries are now focusing on the highly decarbonized heating system i.e. high renewable heating fractions (RHF) in their existing fuel supply, so to restrict fossil fuel boiler use to the minimum possible. Solar thermal systems for steam generation offer a cost-effective way of producing heat with the lowest CO₂ emissions per kWh [14]. However, these systems typically suffer from a lack of steam storage technologies, which limits the solar fraction to below 50% in typical industrial settings [15]. The best economic returns on solar thermal investments are obtained when the heat produced is directly used in the system without the need for storage. Currently, commercially implemented solar thermal steam systems use low-volume storage tanks or synthetic oils as storage media, which can be expensive due to the requirement for pressurized tanks [16], [17]. The cost of storage is also limited by the maximum charging temperature that solar thermal collectors can achieve, which is typically below 250 °C for most commercial collectors used for process heating.

An alternative method for generating renewable heating is using photovoltaic (PV) direct heating for steam generation. Although PV has a lower efficiency than solar thermal systems, resulting in reduced economic feasibility and land use efficiency, it offers some unique advantages. For example, resistance heaters powered by PV can achieve higher fluid temperatures, enabling the charging of high-temperature, low-cost thermal energy storage (TES) systems. Thus, PV + TES system is especially valuable for meeting heating demands during non-sunny hours due to the possible use of low cost storage systems. The higher temperature from PV-driven heating increases the economic feasibility of storage due to a large effective temperature difference in the storage medium, and the possibility of using cheap and abundant materials such as sand or stones as storage media [18].

This paper is based on the premise of technology combination, where individual technology can be utilised to its stronghold, resulting in a final system that is economically feasible and provides high RHF. The technologies employed in such a system complement each other both energetically and economically, resulting in a value proposition that is particularly attractive for industrial applications seeking to achieve high RHF. Accordingly, this paper presents a techno-economic analysis of heat generation using a case study that characterizes industrial boundary conditions requiring steam at a medium temperature range.

The focus of the analysis is on the use of two parallel systems to generate steam and achieve high RHF. Specifically, the technologies analyzed are: (a) a parabolic trough collector with water-based pressurized storage (PTC + TES_water), and (b) photovoltaic panels combined with high-temperature thermal energy storage using sand as the storage medium (PV + TES_sand). The PTC + TES_water system is used to achieve limited solar fraction during the daytime to restrict the heating cost, while PV + TES_sand provides heating during night-time use.

Few studies have analysed hybrid system concepts primarily focused on their applications in electricity generation using concentrated solar power. Consequently, the hybridization approach and related technologies have not been thoroughly studied in the context of process heating [19], [20]. To the best of our knowledge, no prior studies have conducted the techno-economic analysis for the use of the proposed technological combinations for industrial steam generation. As such, the present paper aims to fill this research gap and provide a pathway for industries to adopt this novel system. By analysing the techno-economic feasibility of these systems, the aim is to demonstrate their potential to significantly increase RHF while maintaining cost competitiveness. The specific objectives achieved in this study are as follows.

•
Techno-economic assessment is carried out for sub-systems i.e PTC + TES_water and PV + TES_sand with several constrains by taking in account meteorological, and load parameters. A comparison is made to compare these technologies individually.
•
A hybrid system is proposed combining the technologies, and a system design optimisation is conducted to reach the best hybrid system configuration.
•
The implications of hybrid system on cost of heating, and the land area savings are evaluated.

The likely benefit of system combination is high energetic and economic performance compared to when these technologies are used individually while providing high RHF for industries. This research can have important implications for the industrial sector, as it offers a new perspective on how renewable energy technologies can be integrated to meet their energy needs in a sustainable manner. The sections ahead provide the important backgrounds of technologies in focus, followed by detailed objectives, methodology, results, and discussion.

2. Background

2.1. Solar heating for industrial processes

Solar thermal (ST) collectors are widely utilized to generate heat in various sectors including residential, commercial, and industrial. These collectors have undergone significant development and commercialization over the past few decades. As of 2021, solar heating ranks as the third largest renewable energy source globally, following solar photovoltaic and wind energy [21]. Out of various ST technologies, flat plate and evacuated tube collectors are predominantly used in residential and commercial sectors, for temperatures below 60 °C. For temperatures between 60 °C and 120 °C such as in district heating, double-glazed flat plate collector, or compound parabolic collector can be used [22]. There are few demonstrations and studies on the combined use of flat plate and parabolic trough collector (PTC) for district heating applications [23]. Large scale concentrating collectors used for power generation can reach significantly higher temperature (upto 550 °C), however their use for industrial heat application remains limited due to high costs.

Solar heating for industrial applications (SHIP) is a sub-sector in the solar thermal World, where the collectors are used to generate heating for industries in form of hot water, steam, hot air, or other heat transfer fluids. The temperature range in industrial sectors is very broad (50 °C −800 °C), however, a significant portion of industries use their boilers to generate steam below 200 °C. The steam is then distributed to heat-exchanging elements for process use. IEA solar heating and cooling task 33, 49 and 64 has created a huge knowledge base for the integration of ST collectors for industrial use [24], [25]. Concentrating collectors such as PTC [26], [27], Linear Fresnel reflectors [28], [29], Fresnel lens systems [30] are commercially used technologies for SHIP. There is the possibility to use non-concentrating technologies such as vacuum flat plate collectors [31] or high-performance evacuated tube collectors for industries [32]. SHIP sector is seeing rapid growth in past few years, however, most of the potential is still untapped [33].

Major multinationals have recently shown interest in the installation of ST systems. For example, Heineken and Engie have signed an agreement for a 30 MW solar heating plant for steam generation [34]. Other multinationals are following the trend with pilot or large installations going on Worldwide [35], [36]. In 2021, the total installed collector area for SHIP systems was close to 1.3 million m², with over 1000 recorded installations. According to the World Energy Outlook, the SHIP area needs to increase to 1272 million m² by 2030 for a 1.5 °C global temperature scenario [37].

2.2. Dynamics of SHIP systems

Solar thermal collectors can be integrated into existing boiler systems at various points. The steam generated from these collectors can be used to cover a large portion of the process heat demand by feeding it into the existing steam network. Depending on the technology and temperature levels, steam can be generated using different methods such as direct or indirect steam generation. When 100% of the heat generated by a collector is used to fulfill the load, maximum economic gains are achieved. However, as the collector area increases, the solar production may exceed the heating load [38]. In such cases, the excess heating can either be stored in a storage tank or spilled. In the absence of storage, the increase in collector area will result in a non-linear increase in solar fraction (SF), which in turn leads to a decrease in solar heat utilization. An exemplification is shown in Fig. 1 for a constant load profile, where the levelised cost of heating (LCOH) increases significantly at high solar fraction. Therefore, there exist an optimal solar fraction at which LCOH of solar heating is competitive. Other than the variation of production and demand, the optimal SF can be further restricted by the existing fuel price used in the boiler and the customer’s expectation of economic indicators etc.

The solar heat can be stored, whenever the production is more than the heating load. If the storage is used, the final heating cost from ST system also depends on the LCOH of stored heat, which is further governed by type of storage technology, energy storage density in the installed system, number of cycles per year, capex, and O&M costs. Water tanks (atmospheric or pressurised) are the most common and commercially available heat storage for solar thermal applications typically below 200 °C. Steam storage is non-practical due to the enormously high volumes needed, therefore mostly the storage is done in form of pressurised water (e.g Ruths type storage, or constant pressure tank) or using thermic oil [17]. If the water is used as storage media, the tanks are kept at higher than operational pressure than saturation so to avoid the boiling of water. The heated water is then converted to steam by pressure reduction (flashing) during non-sunny hours. The maximum charging temperature in the tank is restricted by the maximum collector operational temperature. Whereas the minimum temperature in storage is governed by the steam temperature.

Therefore, if a collector operates at a maximum 200 °C and the system need to generate steam at 160 °C, then the effective dT across tank will be 40 °C without any HX. The commercial pressurised tank costs between 1000 and 3000 €/m³, and this high cost usually result in high LCOH of the system compared to when no storage is used. Therefore, to minimize the cost of solar heating systems, an optimal storage volume needs to be achieved, which does not significantly increase the LCOH. However, as the collector area increases, the mismatch between the collector production and load also increases, affecting the energy storage density of the tank and causing LCOH to rapidly increase beyond a certain solar fraction. As a result, an optimal solar fraction exists for specific meteorological characteristics and load profiles, where the LCOH is reasonable. The optimal solar fraction can range from 5% to 90% depending on the heat load profile, with an optimal range of 15%-35% for process heating load that operates 24*7. This limited solar fraction implies that the existing boiler will continue to fulfill the remaining process demand that cannot be met by the ST system, resulting in a limited renewable heating fraction (RHF) for the overall system. Another factor for limited RHF is lack of available land/roof space available in industries for collector’s installation.

To achieve high RHF, a low-cost storage would be needed, which can be charged using excess heat from ST collector. Many Phase change materials (PCM) are under development for steam use, however, are not commercially available [39]. Solid sensible storage (i.e sand, or stone storage) systems can reach very high temperatures without the need for pressurisation, however, their use with solar thermal collectors is restricted. As solid storage media typically have lower heat capacity than water, achieving a high temperature difference in the storage is crucial for obtaining an economic advantage over water-based systems. However, this is difficult to achieve with ST collectors due to their limited temperatures [40].

2.3. Photovoltaic with high temperature thermal storage

To achieve a high RHF, one possibility is to use a parallel system to fulfil the heating fraction not served by ST and medium temperature water-based TES. A technology combination that allows for such a possibility is the use of PV modules to power a resistive element, which can charge a high-temperature TES using cheap materials such as sand or stones. Such storage can operate at atmospheric pressure and can be built using abundant and non-strategic minerals and resources. This storage can be charged using PV or renewable electricity to temperatures between 600 °C and 800 °C with or without heat exchanger, and then discharged for various steam applications, allowing for high ΔT, high energy density, and an economically attractive heat storage option compared to pressurized water/oil storage systems. All these factors allow for a flatter LCOH curve for PV-TES systems even at high solar fractions.

There are many variants of high temp TES demonstrated in research and commercial applications. A preliminary CFD modelling study and experimental evaluation of the rock bed thermal energy storage for lower temperatures is conducted in [41] which resulted in good thermal performance. From a reliability perspective, the studied system has operated over 3 years and 3500 operating hours without any failure. A system with horizontal flow through a packed bed was described in [42]. More recently, Eco-Tech Ceram has demonstrated a mobile horizontal flow waste heat recovery with a 2 MWh capacity and maximum temperature of 600 °C [43]. Also, Siemens-Gamesa has demonstrated a system with a 130 MWh storage capacity with a 5.4 MW electrical heater to charge the system [44]. Stiesdal Storage Technologies have developed a power-heat-power system under the GridScale using hot and cold thermal storages based on rock beds [45]. Polar Night Energy has announced that they began using thermal energy storage based on sand in Finland in 2022 [46]. The system is claimed to have an 8 MWh capacity with nominal power of 100 kW and is designed for providing heat to a district heating network. Other systems, such as Project Malta in USA will test using both a high-temperature molten salt reservoir and an actively cooled reservoir to increase efficiency, with the hope of commercializing the technology [47]. Recent focus on thermal energy storage has been either on temperatures below 100 °C for district heating [48] or for waste heat recovery and power-to-power applications while the medium to high industrial temperature range with heat as desired output has seen less research focus.

High temperature solid storage systems are typically charged using grid electricity, taking advantage of green energy at lower prices. However, in many locations around the world, this may not be possible due to high CO₂ emission index and high grid prices. Additionally, many industries have contractual power purchase agreements with utilities that provide fixed electricity prices over the years. In such cases, TES can be charged using electricity generated from PV, which is a renewable source of energy. One critical issue with generating heat from PV is the lower efficiency compared to solar thermal collectors. Studies have shown that solar thermal can produce 2–4 times more heating output than PV for a fixed land area. As a result, using PV for direct heating, with or without TES, is less land-efficient, even though the LCOH is comparable [49]. However, PV + TES_sand can fulfil the vacuum of thermal storage development in solar thermal systems. A system combination can exist, where solar thermal and pressurised storage systems are designed for optimal solar fraction for any industrial load, and rest of the heating can be met using PV + TES_sand combination. Such system combination is analysed in this study.

3. System design and methodology

The main objective of this paper is to conduct a techno-economic assessment of a novel combined heating system designed to achieve high renewable heating fractions for industries. In this analyzed system, an optimal size of a PTC and pressurized water thermal energy storage will be used to generate steam at the most economical heating cost, up to a limited solar fraction. The remaining heating demand will be fulfilled by a high-temperature TES that uses sand as a storage medium. The TES is charged by photovoltaic (PV) during the daytime and discharged when the PTC + TES_water system is unable to meet the heating demand. The backup heating system, which is an existing boiler, is only utilized when the optimal sizes of both PTC + TES_water and PV + TES_sand are insufficient to meet the heating demand. The analyzed system configuration is presented in Fig. 2.

To achieve the objectives of this study, annual simulations are carried out on both component and system level. A transient system simulation tool (TRNSYS V.18) is used to model the reference system and to calculate the annual output for solar technologies for both PV and PTC [50], [51]. In the next step, the sub-system models are developed, where the interaction of the collector with TES and load is simulated using Python. The economic calculations are done using Microsoft Excel [52].

In specific, the following systems are simulated, and their performance is compared:

Case 1: Reference case (Only boiler without any solar and storage technology).

Case 2: Boiler + PTC + TES_water.

Case 3: Boiler + PV + TES_sand.

Case 4: Boiler + PTC + TES_water + PV + TES_sand.

The method flow chart is shown in Fig. 3. Firstly, the boundary conditions for the reference system (case 1) are defined. This reference system comprises an industrial load profile with steam requirements met by the existing natural gas boiler. The boiler system is modeled and an economic analysis is conducted to determine the LCOH of the current system. Next, the reference system is evaluated using different solar technologies separately (case 2 and 3) and in combination (case 4).

Case 1 is followed by the modeling of the PTC system to find an optimal collector area and volume of medium-temperature pressurised water-based thermal storage to be used in combination with the boiler. The LCOH from the complete system (Case 2) is compared with the reference case. In the next step, PV + TES_sand is optimised based on several iterations of PV and TES capacities. The optimised combination of PV + TES_sand is then combined with a reference case for evaluation (case 3). The final system (case 4) combines all the technologies to compare the LCOH with case 1 to 3. The next sections detail the system boundary conditions.

4. System boundary conditions

4.1. Existing boiler system

The case study used for the reference system is based on data from a real industry in the beverage sector located in Seville, Spain. The facility relies heavily on heating, with 80% of its total energy use dedicated to this purpose, while the remaining 20% is consumed by electricity. The heating is required for various production processes such as pasteurization, distillation, syrup heating, and cleaning in place. Currently, the facility’s heat generation system consists of a natural gas (NG) boiler that generates saturated steam at 158 °C and 5 bar gauge pressure. The steam is then fed to the central manifold and distributed to various processes using a steam piping network. Some of the heat is recovered from the processes as condensate at 70 °C in a condensate tank. This condensate is then degassed at 101 °C along with the required make-up water in the deaerator, using steam injection. The water after de-aeration is sent to the boiler after pre-heating from economizer which heat from flue gases. The water is then turned back into steam by the boilers. The temperatures in the boiler networks are represented in Fig. 4.

An hourly heat load profile is available for a normal operational year. The heat load is measured at the boiler outlet using a steam flow rate for each hour. The variation in the heat load can be attributed to the seasonality of various processes in the facility. The annual heat load is 30,238 MWh, with a peak heating rate of 10.2 MW. The hourly variation in the load is shown in Fig. 5. The facility operates for 5500 h/year, as can be seen in load duration curve in Fig. 6. The average load during operational hours is 5.2 MW. The boiler capacity of 15 MW is used, and is controlled to ensure a fixed temperature of 158 °C at the outlet, by regulating the natural gas flow rate at the burner inlet. One more boiler is integrated in parallel for redundancy purposes and for future expansions. The rated boiler efficiency is 80% when operated at full capacity.

The boiler is modelled in TRNSYS using Type 659. The boiler controls are defined to obtain the set temperature of 158 °C out from boiler. The heat load is given as an input to the boiler to adjust the heating capacity for each hour of the year. Boilers are specified at a rated efficiency, however, there are several factors that affect the boiler performance in a real operating conditions. For example, the boiler is modelled to account for the performance variation with the plant size ratio (PSR). The PSR is a ratio between the maximum output of the boiler defined by peak capacity to the current output. It’s a measurement of how oversized the system is for the current load. As the heat losses from the boiler remain the same at any operational capacity and given temperature, the efficiency of the boiler at reduced capacities are usually lower than the rated efficiency. Eq. (1) is used to determine the efficiency with the variation of PSR(1) $E_{PSR} = - 0.0005 P S R^{3} + 0.0075 P S R^{2} - 0.044 P S R + 1.0222$

Where E_PSR is the fraction of efficiency at variousPSR, and thus E_PSR of 1 would represents that boiler operates at its rated efficiency. The boiler uses Natural gas as fuel, with calorific value of 35.5 MJ/Nm³.

4.2. Parabolic trough collector system

A parabolic trough collector (PTC) is considered as suitable solar thermal technology to generate steam at the required temperature range. The PTC type used for simulations in this study is manufactured by Swedish manufacturer [26]. The focus product is a medium-sized parabolic concentrator specifically designed for industrial steam or hot water applications as shown in Fig. 7. The product is certified by solar Keymark. The three main components of the collector include a) protective glass b) receiver tube c) reflector. The reflector reflects the incoming radiation on to the receiver. The radiation absorbed by the receiver is converted into heat and is then dissipated by the fluid pumped inside the receiver tube. The protective glass helps avoid heat losses and protect the collector from dust, snow etc.

The simulations are done in 2 steps. In the first step, the collector is modelled without any load connected to it. This can be considered as if the collector operates under infinite load, and thus all the heat generated by the collector is fully utilized. The simulations are done using TRNSYS, using parameterisation based on solar Keymark values based on aperture area, as shown in Table 1.

Table 1. Performance parameters for PTC collector.

Peak collector efficiency based on beam irradiance	76.4%
a₁ Heat loss coefficient [W/m²K]	0.80
Incidence angle modifier for diffuse solar radiation	0.12
Azimuth	0 (South facing)
Tracking	Single-axis (N-S tracking axis)

After the collector component simulation, the outputs are used for the system simulations. This is done using a model developed in Python. The model calculates all the energy flows of solar integration corresponding to the weather condition, the heat demand, the solar thermal system layout, and the process physical boundaries (such as operating temperatures and pressures). The tool dynamically simulates the collector interaction with the load, and several iterations are performed to obtain the collector area, and storage volume required to reach an optimal LCOH, and solar fraction. After an optimal field configuration (which includes PTC area, and storage tank volume) is simulated, the net heating load which is not met by PTC collector system is met by the boiler system.

Fig. 8 shows a conceptual schematic of the solar field integration for steam application. There are mainly two loops in the complete system a) Solar field and storage loop b) User loop. The fluids in the two circuit are hydraulically separated via heat exchanger and thus does not intermix with each other. The working fluid in solar and TES loop is water/Glycol mix (40%) to avoid freezing. The user loop consists of same fluid as in boiler network (i.e treated water). The heat is transferred from one loop to another in the steam generator which includes a heat exchanger carrying the solar collector heat carrier and the transfer the heat to water coming from the user to be heated and turned into steam.

To integrate solar thermal energy at boiler level, the collectors field operates parallelly to the boilers: it receives a part (or all) of the boilers’ feedwater in input and it outputs steam to be sent back to the plant steam distribution system. The steam produced by the solar field has properties to the boilers steam, however it has slightly higher temperature and pressure (to have flow priority into the system). Starting from the boiler room on user side, the feed water line (at 100 °C) is tapped and is sent to the steam generator to produce steam at 165 °C. When there is enough pressure in the steam generator, the steam is allowed to fed into the central manifold using control valve. The solar-generated steam has a priority over the boilers-generated steam due to a small overpressure. The saturated liquid separated from the steam in the steam generator is mixed with the incoming water from the feed water and re-heated by solar collectors via heat exchanger. The separation of steam and water, which occurs by gravity, creates two regions (one of steam and one of liquid) in steam generators, and the volumes of which can fluctuate and absorb small power surges or reductions.

Heat carrier enters the solar field, and the flow is controlled by a variable frequency pump and by a recirculation loop valve. The heat carrier is then recirculated into the collectors where it is heated up by the available solar energy. The more solar power is available, the more water can be redirected from the feed water to the steam generator rather than to the boilers. If the solar power production is such that the entire water flow could be handled by the steam generator instead of the boilers and there is excess power that must be dealt with diverting the heat to thermal storage tank for charge. If the heat load is fully met, and the storage is also fully charged, then the collectors are de-focused so not to produce any further thermal output. If the heat load is fully met by solar field + TES_water, then the boilers are put in standby mode. The steam buffer in solar system is typically sized so it can provide the steam supply for time required for the boiler to switch from standby mode to operational mode. The time varies from 2 to 3 min.

Under designed conditions for this study, the heat carrier enters the solar field at 165 °C and exits at 200 °C. The solar circuit is pressurized at a minimum of 20 bar_g to have a safe operational margin before boiling. The storage considered is a pressurized water storage tank with an effective temperature difference of 35 °C. The storage is integrated across collector circuit. Storage is charged to maximum 200 °C and similarly discharged to a minimum 165 °C. The storage is also pressurised to 20 bar g, to avoid steam flashing. Pressure in the tank and solar loop is usually maintained by the nitrogen system. To calculate the heat losses, a cladded rockwool insulation of 125 mm thickness, with effective U value of 0.03 W/m.K is considered.

The pressure drops are calculated based on the pressure drop curve for the collector while assuming specific numbers of bends, T pieces, and valves for fixed pipe length. The electricity consumption of the pumps, trackers, and control equipment is used as operational expenses while calculating the LCOH of the total system. It is assumed that there is negligible piping between steam generator and boiler steam manifold. Therefore, the cost and heat losses are neglected for simplification.

4.3. PV high temperature sand storage system

The sub-system that models the charging of the high temperature sand storage (TES_sand) using PV modules is implemented in Python. Firstly, the PV component is simulated in TRNSYS using typical meteorological year data for the weather station in Seville. The hourly output of the PV system (normalised to kW peak) and the heating load are used as inputs in the Python model. In the hybrid system, the PV field is only used to interact with the thermal storage, without any direct interaction with the load. Thus, all power generated from the PV field is initially used to charge the sand TES. However, in a system, where the PTC system is not in use, then the PV electricity can be used to generate steam directly for the user, with the excess being stored in the sand storage. This control enables a comparison of the LCOH of the PV-TES_sand system with that of the hybrid system.

PV module efficiency in STC conditions is considered at 18%, with an inverter efficiency of 95 %. The PV tilt angle considered is 33°toward the South based on optimal tilt [53]. The specifications of PV module parameters are shown in Table 2.

Table 2. Performance parameters for PV module.

Gross collector area	1.96	m²
Number of PV cells (monocrystalline)	72
Rated power	365	W peak
Electric efficiency at STC	18%
Temperature coefficient of power	−0.41	%/°C
Inverter efficiency	95%
Tilt at angle for PV	33	Degrees
Output de-gradation	0.25%	% per year

The sand based storage considered for this study is based on a commercial product. The storage product is demonstrated for two functional full-scale heat storage systems, both in Finland. The first pilot plant was built in Tampere in 2019 with capacity of 3 MWh. Since summer 2022 a commercial facility with 8 MWh capacity has been in operation in Kankaanpää, designed and built for the customer Vatajankosken Sähkö to produce district heating. In these two facilities, the storage medium is (low-grade) sand, and further research efforts regarding other storage medias is made. The pilots enable testing, validation and optimization of the heat storage solution, and to find the relevant input parameters for simulations in this study.

The heat storage considered is a silo structure filled with storage medium and the energy is stored as sensible heat of the medium. The design allows the average temperature of the medium to reach about 600 °C when the storage is fully charged. The medium is suitable solid granular material, for example low-grade sand. Both the charging and the discharging of the storage are done with a piping system circulating normal pressure air in a closed loop through the storage. The pipes are inside the storage medium so that the air and the storage medium are not in direct contact, and the storage medium is completely static. In the charging mode, illustrated in Fig. 9, the air is first heated up via resistive heaters, after which the hot air enters the pipe system inside the storage medium, releases its heat to it via the pipe walls and comes out to be heated up again. Industrial fans are responsible for controlling the air flow. In the discharge mode, the resistors are off, and the heat is instead transferred from the storage medium to the circulating air. The hot air exiting the storage will enter a heat exchanger, where the heat is transferred to the fluid for steam generation. The air continues from the heat exchanger back to the storage to be heated again when passing by the medium. The storage can control the heat output via controlling the flow of air. Controlling the air temperature is also possible by adjusting the air flow. Silos are created using steel structures with limited thickness, as the whole system does not need pressurisation.

The energy capacity contained in a volume V with average temperature T is given by Eq. (1).(1) $E = ρ V c_{p} (T - T_{ref})$

Here the parameters $ρ$

and $c_{p}$ are the density and the specific heat capacity of the storage medium. For the modelled storage, the properties of low grade sand are use ( $ρ$ = 1700 kg/m³ and $c_{p}$ = 830 J/(kg.K). $T$ refers to maximum charging temperature of the storage. By choosing a reference temperature $T_{ref}$ at which the useful heat transfer essentially halts, the amount of energy stored per one cubic meter of storage medium is calculated. When the material parameters of the storage medium and the values for the maximal temperature and the reference temperature are fixed, the energy capacity of the storage depends only on its volume. The storage is charged at a maximum 600 °C. A pinch temperature of 125 °C is assumed across the heat exchanger to transfer heat from air to user fluid, resulting in $T_{ref}$

of 285 °C (160 °C steam + 125 °C pinch). This results in effective temperature difference of 315 °C in the sand storage, which is 9 times higher for water based pressurised storage in PTC system. The storage is modelled using 2 nodes operating between two layers at charging and discharging temperature.

The comparison of TES water and TES sand for designed boundary conditions is shown in Table 3. The volumetric energy density of TES_sand is upto 3x compare to TES_water. This is mainly due to the high temperature difference in sand storage system. On energy basis, the TES_sand is usually 3–4 times cheaper than an equivalent capacity of TES_water. Therefore, the low cost and high storage capacities allow to have flatter LCOH curve with increasing SF for sand storage system.

Table 3. Comparison of volumetric energy density for analysed TES systems.

Parameter	Value		Units
Storage type	Pressurised Water	Sand based storage	Empty Cell
Designed maximum temperature	200	600	^oC
Designed minimum temperature	165	285	^oC
Temperature difference	35	315	^oC
Specific heat	4.45	0.83	kJ/kg.K
Density	876	1700	kg/m³
Volumetric storage capacity	38	123	kWh/m³

Hourly output of PV, and heat demand to serve is used an input in the model. Storage is assumed to be a cylinder-like structure constructed above the ground with height of about 6 m and with adjustable diameter. Thus, the model adjusts the diameter of the storage, to reach the desired capacity of simulated storage. The investment inputs in model include the storage structure, electrical resistors, storage’s internal heat exchange system, heat exchanger (for connecting the storage to the customer’s heating system), automation system and other essential components like blower and sensors. The connection work to the outer energy grids is not considered for simplifications.

Several iterations of PV and storage capacity are done to find the optimal size, which provides the lowest LCOH. The model assumes that all PV production is put to the storage if there is storage loading capacity left, and similarly all the needed heat is taken from the storage if there is power and energy capacity to be used. If there is excess PV production that cannot be stored, it is possible to feed into the grid and some income may be achieved. For the current study, the monetary value for exported power is not considered, as it has no implications on the cost of heating. All the needed heat cannot be taken from the storage, the deficit is covered by the existing boiler. The investment costs of the PV field and the storage are allocated, with the desired interest rate, to all heat energy taken from the storage during the lifetime of the system.

4.4. Meteorological data

For PTC and PV simulations, weather data based on Typical Meteorological Year (TMY) is used. A TMY is a collation of selected weather data for a specific location, listing hourly values of solar radiation and meteorological elements for one year. The values are generated with Meteonorm v8.1.1, which relies both on weather stations and satellite data [54].

The data in Meteonorm is selected from the latest 20-year period to present the range of weather phenomena for the location in question, while still giving annual averages that are consistent with the long-term averages for the same location. For this project, the weather file for the following co-ordinates is derived.

• Latitude: 37.3^o.

• Longitude: −5.9^o.

• Location: Seville, Spain.

The annual DNI for the analysed location is 2080 kWh/m². and the annual GHI is 2314 kWh/m². The monthly variation in the DNI and GHI is shown in Fig. 10.

Fig. 10. Variation of DNI and GHI for analysed locations.

4.5. Economic input boundaries

The evaluation is done using LCOH as an economic indicator. which is a comprehensive term used to compare different energy systems. It includes system investment costs, fixed and variable O&M costs, and discount rate. The LCOH is calculated using Eq. (2).(2) $LCO H_{PV / P T C} = \frac{CAPE X_{sys} + P r i c e_{el} Â \cdot \sum_{n = 1}^{N} (\frac{W_{Sys}}{{(1 + D R)}^{n}}) + \sum_{n = 1}^{N} (\frac{OPE X_{f + v}}{{(1 + D R)}^{n}})}{\sum_{n = 1}^{N} (\frac{Q_{Yield}}{{(1 + D R)}^{n}})}$

$LCO H_{PV / P T C} = \frac{CAPE X_{sys} + P r i c e_{el} Â \cdot \sum_{n = 1}^{N} (\frac{W_{Sys}}{{(1 + D R)}^{n}}) + \sum_{n = 1}^{N} (\frac{OPE X_{f + v}}{{(1 + D R)}^{n}})}{\sum_{n = 1}^{N} (\frac{Q_{Yield}}{{(1 + D R)}^{n}})}$ $LCO H_{PV / P T C} = \frac{CAPE X_{sys} + P r i c e_{el} Â \cdot \sum_{n = 1}^{N} (\frac{W_{Sys}}{{(1 + D R)}^{n}}) + \sum_{n = 1}^{N} (\frac{OPE X_{f + v}}{{(1 + D R)}^{n}})}{\sum_{n = 1}^{N} (\frac{Q_{Yield}}{{(1 + D R)}^{n}})}$ $LCO H_{PV / P T C} = \frac{CAPE X_{sys} + P r i c e_{el} Â \cdot \sum_{n = 1}^{N} (\frac{W_{Sys}}{{(1 + D R)}^{n}}) + \sum_{n = 1}^{N} (\frac{OPE X_{f + v}}{{(1 + D R)}^{n}})}{\sum_{n = 1}^{N} (\frac{Q_{Yield}}{{(1 + D R)}^{n}})}$ $LCO H_{PV / P T C} = \frac{CAPE X_{sys} + P r i c e_{el} Â \cdot \sum_{n = 1}^{N} (\frac{W_{Sys}}{{(1 + D R)}^{n}}) + \sum_{n = 1}^{N} (\frac{OPE X_{f + v}}{{(1 + D R)}^{n}})}{\sum_{n = 1}^{N} (\frac{Q_{Yield}}{{(1 + D R)}^{n}})}$

Where:

= Capital cost of complete system including installation and commissioning.

= Operation cost of complete system including fixed and variable O&M.

Unit price of electricity.

= Annual power consumption of system accessories.

DR = Discount rate [%].

N = Project lifetime [years].

= Thermal demand met by the system under study.

The input data for capital and O&M expenses data for individual technologies is provided by the manufacturer based on their experience. The fuel cost for existing boiler is obtained from the user. The LCOH is evaluated for 15 years’ time frame, and at 0 % discount rate. Even though the chosen technologies have a longer technical lifetime, the LCOH is capped to a 15-year timeframe as based on experience, many industrial users are often not interested in making investments for long-term horizons. The input parameters used for simulations are shown in Table 4.

Table 4. Economic parameters used for analysis.

Empty Cell	Empty Cell	Empty Cell	Units	Remarks
Boiler	Fuel type	Natural Gas
Empty Cell	Boiler nominal efficiency	80%
Empty Cell	Levelized fuel price over 15 years	100	€/MWh	Includes boiler efficiency, O&M costs, degradation.
Empty Cell	Electricity price	150	€/MWh
PTC + TES_water [35]	Collector cost	200	€/m²
Empty Cell	Balance of plant cost	100	€/m²	Steam generator, integration, piping, installation & commissioning
Empty Cell	Thermal storage capex (TES_water)	1500	€/m³
Empty Cell	O&M cost	1%		Percentage of capex per year for both collectors and storage
Empty Cell	Land area index	2.3	m²	land area needed for 1 m² collector area to avoid row shading
PV + TES_sand [46]	PV capex	800	€/kW peak	Including installation and commissioning
Empty Cell	O&M cost PV	1 %		of capex per year
Empty Cell	Efficiency for resistive heater	99 %
Empty Cell	Land area required PV	12	m²/kW peak
Empty Cell	TES capex	25 000	€/MWh	including installation and commissioning. MWh calculated at designed conditions with 315 °C temperature difference.
Empty Cell	TES O&M	2 %		of capex per year

5. Results

5.1. PTC simulation results

The performance of the PTC collector is evaluated for a range of collector field areas and storage tank volumes. The annual thermal output of the PTC collector at designed conditions is 600 kWh/m², when all the heat is utilised in the user loop. The simulation considers a collector field of up to 100 000 m² and a storage volume of up to 10 000 m³ for the given load profile, and the results are analyzed. Fig. 11 illustrates the effect of increasing the collector area on the heat absorbed by the user. The results indicate that up to a collector area of 15 000 m², most of the heat generated by the solar field is directly supplied to the load without any need for thermal storage. This is due to the heating demand during daytime, which can be met when the collectors are producing heat. The effect of storage is more prominent at higher collector areas, where it is needed to serve the heating demand during non-sunny hours. Increasing the collector area without storage results in a marginal increase in the absorbed heat.

Fig. 11. Variation of absorbed heat with collector area and tank volume.

For each field size, the storage volume is optimized to limit the increase in LCOH. Therefore, some of the heat stored by the solar heat is allowed to spill to avoid sizing the storage for peak production. This also avoids an exponential increase in storage size and thus in LCOH. However, this control effect also increases excess heat, which cannot be delivered to the user under the given load profile, as shown in Fig. 12. The economic value of this wasted heat is not considered for the analysis.

Fig. 12. Variation of wasted heat with collector area and tank volume.

The effect of solar fraction with increasing collector area is shown in Fig. 13. SF increase linearly upto collector areas when no heat is wasted. Afterward, due to diminished returns on the heat delivered, the SF curves flattened at high collector areas. For the simulated collector and storage range, solar fraction upto 78 % is obtained. It is possible to simulate higher solar fractions, but it is restricted due to the computational effort required.

Fig. 13. Variation of solar fraction with collector area and tank volume.

Due to the limited heat delivered/m² with increasing collector area, the LCOH also increases, and the trend for the same is shown in Fig. 14. The LCOH ranges from 50 €/MWh to 130 €/MWh for simulated range of the solar fields and tank volume. The least LCOH of PTC is obtained when all the heat generated by the solar field is utilised by the user, which occurs mostly with collector areas below 15 000 m² aperture area.

Fig. 14. Variation of LCOH for analysed range of collector area and tank volume.

The heating demand that is not met by the PTC + TES_water system is fulfilled by the user’s existing boiler. Therefore, it is important to analyze the combined levelized cost of heat for the integrated system, which includes PTC + TES_water + boiler, as this is significant for the user. Fig. 15 shows the LCOH for different solar fractions for PTC + TES_water with and without boiler. At 0 % SF, all heating demand is met by the boiler, resulting in an LCOH of the system at 100 €/MWh. As the PTC + TES_water system can produce heat at a lower cost compared to the boiler, the combined LCOH decreases with increasing SF and reaches its minimum value at 90.1 €/MWh at 43 % SF (i.e., 57 % boiler use). The maximum SF resulting in any decrease in LCOH (compared to boiler only) is 70 %, after which the integration of PTC is not economically beneficial.

Fig. 15. Variation of LCOH with solar fraction for PTC systems with and without boiler.

Results reflects that PTC + TES_water provides the least LCOH at lower solar fractions, and therefore a good technological option. However, the impact on the final cost of heating to the user is limited due to the restricted SF. The solar fraction range between 30% and 50% is found to result in the lowest cost of heating to the user.

5.2. PV driven sand storage system performance

For specified location and load profile, the annual electrical output (including conversion losses) is 1610 kWh/kW peak and the daily variation in the performance is shown in Fig. 16.

The PV electricity generated is converted into heat, which can be used to satisfy the heating demand either directly or through thermal storage. Several iterations are carried to scan the optimal configuration of PV size and TES_sand for a range of solar fractions. The aim is to determine the configuration that can provide the lowest cost of heating to the user at any given solar fraction.

Once the most optimal PV system configuration and thermal storage tank volume are determined, the system is integrated with the user’s existing boiler to analyze the LCOH of the combined system. This analysis is important because the system’s LCOH reflects the total cost to the user to produce a unit of heat over the system’s lifetime. The results of the analysis are presented in Fig. 17, which shows the combined LCOH of the PV + TES_sand system and the boiler for different solar fractions.

Results show that the PV + TES_sand system can provide a minimum LCOH of 73.9 €/MWh. However, the LCOH increases to 75.4 €/MWh at 23 % SF. Similar to PTC system, The system achieves the lowest LCOH when most of the PV heating is directly supplied to the load without the need for storage. There is a high mismatch in solar production and the user demand. At any given SF, it is determined by LCOH, if the excess from PV/PTC should be allowed to spill or stored. In both PV and PTC case, the spillage of electricity and heat up to certain SF is preferred instead of using storage. However, as the mismatch increases significantly after a specific SF, thus increasing the LCOH in direct use case, then use of storage is preferred.

The LCOH of boiler integrated system (PV + TES_sand + boiler) continues to decrease with increasing solar fractions, reaches its lowest value at 87.8 % SF. The results also reflect that PTC is effective in generating heat at lower LCOH compared to PV + TES_sand but only up to a limited SF. The increase in LCOH for PTC + TES_water system is higher due to the thermal storage cost dynamics. Whereas, PV + TES_sand system has flatter LCOH even at higher SF. This results in lower LCOH of PV + TES_sand + boiler system at high SFs, compared to PTC + TES_water + boiler system.

Fig. 18 compare the LCOH for PTC and PV systems without boiler. When storage is integrated into the PV + TES_sand system, the LCOH increases, especially at around 23% solar fraction. However, the increasing effect is not as significant as in the PTC + TES_water system because the storage is effectively utilized due to high temperature differences in storage and the lower costs. The LCOH remains relatively stable up to very high solar fractions, and then increases to 86.2 €/MWh at a solar fraction of 88%. It is challenging to achieve solar fractions of more than 90% at reasonable costs, mostly due to the high-capacity requirements for both PV and storage, which significantly increase the LCOH. The LCOH for both PTC + TES_water and PV + TES system becomes equal at 53% SF (LCOH of 85 €/MWh), and after that PVTES has lower LCOH than PTC TES. The lowest LCOH of PTC will be equal to that of PV if the cost of PTC is increased to 425 €/m², which represents a 40% increase from the assumed cost in the analysis.

5.3. Hybrid system analysis

The lowest LCOH of PTC-TES_water system is obtained at 50 €/MWh, whereas the same for PV-TES_sand is at 73.9 €/MWh. However, the LCOH increase with storage for PTC system is significantly higher compared to the PV system. This bring up the possibility to combined PTC + TES_water and PV + TES_sand system to further decrease the LCOH of overall system, where the aim is to leverage the low LCOH of PTC upto limited solar fraction. To simulate this step, different capacities of PTC + TES_water are combined with PV + TES_sand system so to reach the combined solar fraction of 90%. The solar fraction limit of 90 % is found optimal, as the LCOH increase significantly after this as shown earlier in Fig. 17.

Fig. 19 demonstrates the impact of different PTC system capacities on the overall system LCOH. At 0% SF on the x-axis, which indicates that no PTC system is used, the LCOH of the PV + TES_sand + boiler system is 90 €/MWh. However, as the PTC fraction increases, the LCOH decreases, and the minimum value of 83.5 €/MWh is reached at a PTC + TES_water fraction of 30 %, a PV + TES_sand fraction of 60%, and a boiler fraction of 10 %. At this point, the combined system solar fraction is 90 %. If the PTC SF is increased beyond 30 %, the LCOH of the system starts to increase again. When the PTC + TES_water fraction reaches 63 %, the combined system LCOH is equivalent to the case where no PTC system is used (i.e., at 0% PTC solar fraction). This suggests that the use of PTC systems becomes economically disadvantageous beyond a 63 % SF limit.

In Fig. 18, the PV and PTC LCOH crosses each other at 53% SF. This can be compared in Fig. 19, by looking at when the LCOH of PTC integrated system equals to LCOH without PTC. This point occurs at around 60% PTC SF in Fig. 19. The reason for this increase (from 53% to 60%) is that when PTC is integrated, the daytime demand plus part of night time load is met by PTC system. Which means that PV + TES_sand will have less heat to storage for non-linear reduction in the size. This further increase the LCOH of PVTES system, and the crossing point on LCOH increase slightly in hybrid system.

The hybrid system that achieves 90 % SF at minimal LCOH consist of PTC collector with aperture area of 22 583 m² and TES_water volume of 833 m³, to deliver 9035 MWh to the user. The PV capacity required is 14 MW, and TES_sand capacity of 435 MWh, to deliver 19 092 MWh heat to user. The rest of heat is still delivered by existing boiler. The system size, energy values and LCOH for hybrid system are shown in Table 5.

Table 5. Results for hybrid system to provide 90% SF.

Empty Cell	Parameter	Value	Units
1	PTC + TES_water
1.1	PTC aperture area	22 583	m²
1.2	TES_water volume	833	m³
1.3	Heat delivered to user	9035	MWh/y
1.4	Heat from PTC to user	6681	MWh/y
1.5	Heat from TES to user	2354	MWh/y
1.6	Losses (thermal losses and spillage)	4251	MWh/y
2	PV + TES_sand
2.1	PV capacity	14	MW peak
2.2	TES sand capacity	435	MWh
2.3	Heat from PV to TES_sand	19 092	MWh/y
2.4	Heat losses in thermal storage	892	MWh/y
2.5	Heat from TES_sand to user	18 200	MWh/y
3	Boiler
3.1	Heat from boiler to user	3003	MWh/y
4	LCOH
4.1	LCOH for PTC + TES_water	68.9	€/MWh
4.2	LCOH for PV + TES_sand	89.2	€/MWh
4.3	LCOH boiler	100	€/MWh
4.4	LCOH hybrid system (PTC + TES_water + PV + TES_sand + boiler)	83.5	€/MWh
4.5	LCOH for only PV + TES_sand + boiler system for 90% SF	90	€/MWh

Overall, the combined system has high solar fraction (90 %) and lowest LCOH compared to only PV + TES_sand + Boiler system. This reflects that the combined system is better in terms of energy and economic performance. Although achieving a solar fraction of more than 90% is possible but not economically beneficial for the user.

The use of PTC not only results in lower LCOH, but also in significant land area savings. This advantage is due to the higher conversion efficiency of PTC compared to PV systems. For industrial users with limited land area, this advantage can be especially important. By using PTC + TES_water in a PV + TES_sand + boiler system, significant land savings can be achieved.

The analysis for land savings is made w.r.t a base case, where PV + TES_sand + boiler system is used without any PTC collector (0% PTC solar fraction). For base case, the LCOH is 90 €/MWh, and the land area requirement for PV system (18 MW) is 216 000 m² as shown in Fig. 20.

With increasing PTC fraction, the land area requirement continues to decrease due to higher PTC efficiency. At minimum overall system LCOH of 83.5 €/MWh, the PTC fraction is 30 %, and the land area requirement is 196 230 m², a 10 % decrease in land area w.r.t base case. After 30 % PTC SF, the land area requirement continues to decrease and reaches it’s minimum value at 46 % PTC SF, resulting in 17% less land area usage compared to base case. At this point (46% PTC SF), the LCOH is 90 €/MWh, i.e same as base case without PTC. This represents that instead of just installing PV + TES_sand system, the addition of PTC can reduce the land area use by 17 % without increasing any LCOH. Similarly, land area of 10 % can be saved with PTC, with further decrease of 6 €/MWh in overall system LCOH. This clearly shows the advantage of combining PTC + TES_water, and PV + TES_sand technologies to result in economic and land use benefits. Importantly, the complete system result in high decarbonisation level (90% solar fraction) and helps industries to reduce their emissions.

6. Discussions

Industrial heating decarbonisation presents a significant opportunity that requires attention from various stakeholders. There are several technologies available in the market at different levels of readiness that can assist industries in transitioning away from fossil fuels. Some of the key technologies in this sector include steam heat pumps, solar thermal collectors, PV driven heating, and waste biomass. Each of these technologies has its advantages and disadvantages, and hybrid systems that combine multiple technologies can complement each other to create an effective technology collaboration.

This paper presents the techno-economic assessment of a novel system combination, applied for a food & beverage industry, located in Seville, Spain. In analysed system, Solar thermal collector such as PTC, with pressurised based water storage are used to provide limited solar fraction. A significant portion of heat demand is then met by PV driven electrical heater, which charges a sand based thermal storage upto 600 °C. The existing boiler use is restricted to minimal, only when the heat demand can not be met by PTC-TES/PV-TES system.

The results shows that optimal combination of PTC + TES_water, and PV + TES_sand system, with boiler backup has the lowest LCOH, when compared to if these technologies are used individually. The boiler only system has LCOH of 100 €/MWh. If the PV-TES system is used without any PTC collectors, the LCOH is at 90 €/MWh at 90 % SF, with high land area use. The introduction of PTC, upto limited SF; helps to reduce the LCOH and significant land area savings. The use of PTC resulted in an LCOH decrease of further 6 €/MWh, lowering LCOH to 84 €/MWh, and providing land area savings of 20 000 m². In a combined system, the PTC and PV systems complement each other to achieve better economic and land area performance, while also achieving a very high combined solar fraction.

The analysis conducted in the paper is detailed, but it is limited a fixed set of boundary conditions that were chosen using data from a real industrial plant. The main premise of the paper is that PTC technology can produce heat at a low LCOH, but as the solar fraction increases, the LCOH also increases at a higher rate. On the other hand, the PV + TES_sand system has a higher LCOH at lower SF, but the LCOH curve flattens out as the SF increases, mainly because of the better use of storage. This premise remains true even if the load profiles and temperature boundaries chosen for the analysis are different within a certain range. The important factor affecting the LCOH would be the cost of the system. In the current analysis, the lowest LCOH from both PV (73 €/MWh) and PTC (50 €/MWh) is achieved when all generated heat is directly utilized in the system without thermal storage. The low LCOH of PTC allows for the possibility of hybridization, which has a positive impact on final hybrid system’s heating cost. However, the LCOH of PTC will be equal to that of PV if the cost of PTC is increased to 425 €/m², which represents a 40% increase from the assumed cost in the analysis. In this scenario, the primary reason for using PTC collectors would be land area savings, as no decrease in LCOH will be observed.

The TES cost has a significant effect on the LCOH for PTC system. In the analysis, a fixed storage of 1500 €/m³ is considered based on market prices. However, the there is a possibility for cost reduction at higher volumes, projected upto 1000 €/m³ in [55]. Fig. 21 compare the LCOH for PTC + TES_water vs PV + TES_water without any thermal interaction. At TES_water cost of 1000 €/m³, PTC can provide heat at low LCOH upto 66 % SF compared to PV + TES_sand system.

The land area savings provided by PTC are particularly important for industries with limited space. While PTC collectors can be installed on the roof, it is generally more cost-effective to install them on the ground. PV, on the other hand, can be installed on the roof with standard structures, and if installed horizontally or in an east–west direction, the space requirement per kW of PV can be significantly reduced. This may result in a higher land use index for PV and lower area savings impact of PTC in the overall system. If the land cost is considered in the analysis, and If both PV and PTC are installed on the land, then results will favour PTC to higher SF. It is important to note that in the final hybrid system configuration, all the electricity generated by PV is converted into heat and stored in TES_sand. However, the same is not true for the PTC system, as part of the heat generated by PTC remains unutilized (or spilled) to limit the tank volume and LCOH. If this excess heat can be utilized in an industrial plant for other processes, the overall system will favor PTC and decrease the system’s LCOH even further. An alternative system to PTC in the proposed hybrid system could be the use of a steam heat pump to generate heat during the daytime using PV electricity. Such a system configuration could be interesting from a future study perspective.

The study assumes a steam temperature of 158 °C, and increasing this temperature would result in higher heat losses in the PTC system, which would ultimately increase the LCOH. However, the temperature increase does not affect the efficiency of the PV system. As a result, the optimal PTC SF in the overall system tends to decrease at higher steam temperatures. Any decrease in PV prices for large-scale plants can further help to reduce the LCOH of the combined system.

In this study, the load profile used had a significant daily variation. There is a high mismatch in collector production and demand, which results in high LCOH even at low solar fractions due to spillage of electricity or heat. It is important to note that different industries may have different load profiles, which could affect the results obtained from the analysis. For example, the use of daytime loads can help to further decrease the LCOH, as there will be a higher proportion of direct use heat in the system. Furthermore, the optimal configuration of the hybrid system will depend on the current fuel cost in the boiler. If the fuel price is lower than the LCOH of PV/PTC, then using these technologies will not be economically viable. Conversely, if the fuel price is higher than PTC but lower than PV, then only the use of PTC collectors will make economic sense. Therefore, it is important to consider the current fuel prices when determining the most economically feasible configuration of the hybrid system. To further improve the economic viability of solar thermal technologies, there is a need to reduce the cost of collector by mass-production and standardising for the balance of plant.

The study is limited to one value of existing heating costs for customers, which is averaged over 15 years. However, fuel price volatility is significantly higher and difficult to predict. This uncertainty regarding fuel prices and production costs for industries can be avoided by using renewable heating systems, such as those proposed in this paper, to reach decarbonization goals. The expected increase in CO₂ tax in the coming years can help to promote these system concepts. On a technological level, more innovation in business models, such as heat purchase agreements, is needed to reduce the customer’s risk for implementing new concepts.

The analysis did not consider the comparative indicator of CO₂ emissions. However, it is worth noting that solar thermal technologies have low life cycle emissions in comparison to PV, due to their lower use of rare earth minerals. From a CO₂ emissions standpoint, the use of PTC collectors in the hybrid system could have a greater impact than PV collectors. Therefore, it is important for future studies to consider both economic and environmental factors when determining the optimal configuration of the hybrid system.

7. Conclusions

The main aim of this paper is to do a techno-economic assessment for a novel combined heating system to reach high renewable heating fractions for industries. In the analysed system an optimal size of parabolic trough collector (PTC) and pressurised water TES are used to generate steam at most economical heating cost, up to a limited solar fraction. The part of remaining heating demand is served by a high temperature TES using sand as a storage media, which is charged by PV during the day time, and discharged when PTC + TES_water cannot meet the heating demand. Back up heating system (existing boiler) is used only when optimal sizes of both PTC + TES_water and PV + TES_sand cannot fulfil the heating demand.

The system simulations are done using model developed in TRNSYS, python and Microsof Excel. The case study chosen is based on data from a real food & beverage industrial plant located in Seville, Spain. The key conclusions of the study are as following:

•
The combination of PTC + TES_water and PV + TES_sand with an existing boiler has shown the best economic performance. The overall system achieved the lowest LCOH of 83.5 €/MWh, which is a 16 % decrease compared to the cost of fuel used in the boiler alone. The solar fraction of the combined system is 90 %, with only 10 % of the heating demand still met by the boiler. The PTC + TES_water system contributes to meeting 30 % of the total demand, while the PV + TES_sand system meets 60% of the demand.
•
The use of a PTC collector in the overall system reduces the land area requirement due to its high conversion efficiency. At the minimum LCOH point of 84 €/MWh, the use of a PTC system has resulted in a 10 % land savings compared to a case where no PTC collectors are used. Increasing the PTC fraction shows that land savings of up to 17% can be achieved without increasing the LCOH relative to the base case of no PTC use.
•
On a component level, the PTC system has achieved the lowest LCOH of 50 €/MWh, while the PV system LCOH is 73 €/MWh. However, the increase in LCOH with solar fraction is significantly higher in the PTC system due to the expensive storage required. In contrast, the trend for the PV + TES_sand curve is flatter up to 90% solar fraction.

The combined system has better economic and land use performance compared to when these technologies are used individually. Overall, the system has very high solar fraction of 90%, while still being competitive to existing fuel cost. The findings of this study can help industries evaluate the techno-economic feasibility of adopting renewable heating systems and enable them to make informed decisions about transitioning to sustainable heat sources. The findings also highlight the importance of development needed in medium temperature TES systems for solar thermal applications.

8. Authorship contribution statement

•
Puneet Saini: Conceptualisation, Methodology, Simulations: PV-TES, Analysis, Writing-First and final draft.
•
Ville Kivioja and Liisa Nasakali: Model development, and commercial inputs for sand storage. Review.
•
Carlo Semeraro: Critical comments and commercial inputs for PTC system.
•
Andrea Gambardella: Support with simulations for PTC-TES.
•
Joakim Byström: Supervision, and critical feedback
•
Xingxing Zhang: Supervision, and critical feedback

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

The authors would like to thank Absolicon solar collectors AB and Polar Night Energy Oy for their support to carry this study. Authors also acknowledge support from J Gustav Richert foundation (grant number 2023-00840).

Data availability

Data will be made available on request.

References

[1]
“World Energy Outlook 2022 – Analysis – IEA.” https://www.iea.org/reports/world-energy-outlook-2022

(accessed Apr. 05, 2023).