The design and selection of the yogurt processing unit are largely based on some basic parameters such as energy efficiency, maintenance, especially product life cycle, and environmental impact. Figure 1 shows a solar assisted yogurt processing unit developed to handle raw milk and its fermentation into yogurt timely at the production site. It consisted of a round-shaped fermentation chamber (560 mm diameter and 230 mm depth) made of stainless steel (food grade SS 304) having a capacity of 50 L which was surrounded by a heating coil (3.5 m long, 40 mm wide, and 12.5 mm high). The walls and bottom of the fermentation chamber were insulated with 100-mm-thick PU (polyurethane) material so that heat loss through conduction and convection could be reduced. An electric motor with variable frequency drive (VFD) was installed on the top of the chamber to rotate a stirrer for maintaining uniform temperature in the processed product. For cooling purposes, the bottom surface of the chamber was fabricated by a pillow plate which acted as an evaporator as well. The use of a pillow plate heat exchanger not only reduces the size and cost of the unit but also provides a higher heat transfer coefficient in comparison to the conventional coil heat exchanger. For cooling of yogurt, one ton of rotary compressor compatible with R-410 A (environmentally friendly) gas was installed using an inverter kit to reduce torque load to run on 2 kW_p PV modules.

For heating of raw milk, the yogurt processing unit was connected with a hot water storage tank (100 L capacity) which received heat from a solar evacuated tube collector (2.5 m²) having a connection through polyvinyl chloride (PVC) pipe fittings, as shown in Figure 2.

The outer and inner diameters of ETC tubes were Ø58 mm ± 0.7 mm and Ø47 mm ± 0.7 mm, respectively. The glass tube length was 1800 ± 5 mm, and the vacuum was P < 5 × 10⁻³ Pa. The thermal energy absorbed by the collector was transferred to 100 L of water in the storage tank to raise the water temperature to 90°C. A centrifugal pump (Wilo-SP106) was installed between the hot water storage tank and evacuated tube collector for the circulation of propylene glycol solution (50% by volume). The pump could be operated at three variable speeds (600 L/h, 900 L/h, and 1100 L/h), and it required 80 W power at maximum speed. The current research was conducted at a flow rate of 600 L/h. Heated glycol solution entered the storage tank and transferred its heat to water while passing through the helix-type heat exchanger present in the storage tank. In order to transfer heat from the storage tank to the yogurt processing unit, another water circulation pump (stainless steel centrifugal, WB50/025D, 50 L/min) was installed between the outlet of the hot water storage tank and the inlet of the yogurt processing unit to circulate the hot water through the square spiral coil heat exchanger to raise the temperature of milk up to 80°C.

Being a closed cycle, an expansion vessel (12 L) was also provided to avoid the high pressure build up in the system. The controller turns on the circulation pump (Wilo-SP106) when the temperature differential between the glycol solution leaving the collector and the water in the lower portion of the storage tank exceeds 5°C and turns it off when the differential is below 5°C or when the water temperature of the storage tank exceeds 90°C. An electric heating coil (2 kW) is also mounted inside the hot water storage tank to process the milk/yogurt in the low or no sunshine hours.

Trials were conducted to assess the heating and cooling performances using three different volumes of raw milk (30, 40, and 50 L). Before conducting an experiment, the heating component of the system was turned on to raise the water temperature of the storage tank up to 90°C, which normally took 2–3 h depending on solar radiation. After that, raw milk was poured into the fermentation tank, and the circulation of hot water from the storage tank to the fermentation chamber was turned on. The heating of raw milk up to 80°C took about 140 min, and it was continuously stirred at the speed of 36 rpm. In order to bring down the temperature of heated milk to 43°C, recommended for fermentation of milk, tap water was passed through the heating coil in an open loop controlled by manually operating valves. Inoculation of starter culture was performed (2-3%) at this temperature and was maintained through a solenoid valve which controlled the circulation of water for a period of 5–6 h until the required pH (4.85–4.5) was attained. After that process, the cooling process of yogurt was started by turning on the refrigeration unit to bring down the temperature of yogurt below 8°C to reduce the bacterial activities, which normally took 48–103 min depending on the quantity of the processed milk and stirrer speed. For this, three different speeds (36, 18, and 6 rpm) of stirring were selected using a variable speed motor. Experiments were conducted with three milk volumes (50, 40, and 30 L). Each experiment was conducted three times, and values were averaged. For data acquisition, a controller with resistance temperature detector (RTD)-based temperature sensors (accuracy ± 0.1°C) was used to measure temperature at the inlet and outlet of ETC, top and bottom of the hot water storage tank, and inside the fermentation chamber. A portable pH meter (ML1010) was used to measure the pH of milk during the fermentation process. To assess the performance of the installed PV system, a clamp meter (Fluke 345PQ) and pyranometer (METEON) were used.

There are different options to couple various types of solar collectors for heating milk, like flat plate collectors and solar concentrators, but the evacuated tube collector (ETC) was chosen because of its high efficiency, compactness, and durability and it does not need any tracking system. The glass vacuum tube material was borosilicate glass, and the structure was a concentric dual-tube geometer. The average heat loss coefficient from tubes was 0.8 W/m²K. For the storage of hot water, a stainless steel tank with 100 L storage capacity was installed. The temperature from ETC can be achieved up to 100°C plus. The input power at ETC is calculated by the following equation (Munir et al., 2021): G ′ = ( I t   × A c ) 1000 .                                   (1) where G′ is the rate of incident solar energy at the collector (kW), I _t is total solar irradiance or GHI (global horizontal irradiation) (W/m²), and A_c is collector surface area (m²) and is calculated by multiplying effective tube length (exposed to solar radiation) by its diameter (aperture) and the number of tubes.

The thermal efficiency of the solar evacuated tube collector depends on solar input energy and losses and heat transferred to the working fluid (water + glycol solution), and its value lies in the range of 70–80%. So, the amount of heat energy transferred can be estimated by multiplying the mass flow rate with the specific heat of the fluid and the change in temperature (Munir et al., 2021): Q ′ = m ′ C p ( T 2 − T 1 ) . (2) where Q′ stands for the energy-transferred rate (collector to water–glycol solution) in kW, C_p is specific heat capacity (kJ/kg K), m` stands for flow rate (kg/s), and T₂ and T₁ are the temperatures at respective points shown in Figure 2: η =   Q ′ / G ′ . (3)

In order to transfer heat from the storage tank to the yogurt processing unit, a water circulation pump was installed between the outlet of the hot water storage tank (point 5) and the inlet of the yogurt processing unit (point 6) to circulate the hot water through the square spiral coil heat exchanger to raise the temperature of milk up to 80°C, and its conditions continued to change during the heating process.

In addition to energy analysis, exergy analysis has been used widely in recent years (Aghbashlo et al., 2013). Fundamental principles of energy and mass conservation are used for the assessment of energy flow in and flow out of the system. For this, the following equations are used to calculate exergy flows (inflow, outflow, and loss) (Amjad et al., 2020): Exergy = c p w [ ( T − T a ) − T i ⁡ ln   T T a ] . (8) where T is the temperature of the respective points, T_a is the ambient temperature, and the subscript i is for initial.

For the thermal analysis, the entire system was divided into heating and cooling processes.

The ETC header, hot water storage tank, PVC pipes for connection, and yogurt processing chamber are of cylindrical shape. The required physical and thermal parameters, as tabulated in Table 1, are used to calculate the lateral conduction, convection, and radiation losses from their walls with insulation and without insulation.

For entire calculations, ambient, space temperatures, and solar insolation were supposed to be 25°C, 15°C, and 800 W/m², respectively. For a cylindrical layer that is exposed to a convective heat transfer mechanism on both sides, the steady-state heat transfer (φ) can be calculated as (Cengel, 2004; Munir et al., 2021): ϕ = T i n − T a m b R c o n d + R c o n v . (15) where T _in stands for the inside temperature of the processing chamber, T _amb represents ambient temperature, R _cond symbolizes the resistance due to conduction, and R _conv denotes convection resistance.

Thermal conduction resistance for the lateral cylindrical wall is calculated as (Cengel, 2004; Munir et al., 2021): R c o n d = 1 2 π   λ   L ln   ( r e x t r int ) . (16) where “r _int ” and “r _ext ” are the inner and outer radius of the cylindrical shape of the hot water storage tank, yogurt fermentation chamber, header of ETC and PVC pipes. λ is the material thermal conductivity (stainless steel sheet 15 Wm⁻¹K⁻¹), and L stands for the lateral length of the cylindrical body, and the area of the outer sheet and inner sheets would be equal when r _int is equal to r _ext (in case of no insulation).

The conductive resistance for multilayered plane wall with insulation is calculated by adding the conductive resistance for all layers and is given in Eq. 17: R c o n d = 1 2 π   λ   L [ ln   ( r e x t r int ) i s + ln   ( r e x t r int ) i n s + ln   ( r e x t r int ) o s ] . (17) where subscripts “os,” “ins,” and “is” represent the outer sheet, insulation, and inner sheet, respectively. The values of conductivity resistance for outer sheet, insulation, and inner sheets are calculated for the hot water storage tank, yogurt processing unit, the header of ETC, and insulated PVC pipes and tabulated in Table 2.

The values of conduction resistance for the circular part (top and bottom) of the hot water tank, yogurt processing unit, and header of ETC are calculated by the following relation (Cengel, 2004; Munir et al., 2021): R c o n d = 1 π ( t i s λ i s r i s 2 ) . (18) With insulation, the conductive resistance for the multilayered plane wall is calculated by adding the conductive resistance for all layers and is given in Eq. 19: R c o n d = 1 π ( t i s λ i s r i s 2 + t i n s λ i n s r i n s 2 + t o s λ o s r o s 2 ) . (19) where “t” is the wall thickness, “r” is the radius of circular section, and subscripts “os,” “ins,” and “is” represent the outer sheet, insulation, and the inner sheet, respectively. The values of conductivity resistance for outer sheet, insulation, and inner sheets are calculated and tabulated in Table 2. Convection resistance is calculated by the following relation (Munir et al., 2021): R c o n v = 1 h   A . (20) where ‘h’ stands for the natural convectional coefficient, and its value is assumed. ‘A’ is the area of the sheet. In order to calculate the resistance due to convention, an iterative technique was used by taking values of the physical parameters like air density (ρ _a=1.050), dynamical viscosity (μ)-air (μ _a = 0.000020500), conductivity-air (ʎ_a = 0.0295), air-specific heat capacity (a_a = 1,009), static viscosity-air (V_a = μ_a/ρ _a = 0.0000195,238), and Prandtl (Pr)-air (Pr=C_a* μ_a/ʎ_a = 0.7011).

The Grashof number is calculated as (Munir et al., 2021): G r = β   g   ΔT ρ 2   L 3 μ 2  . (21) where “β” stands for the coefficient of thermal expansion, g stands for gravitational acceleration (9.81 ms⁻²), and “L” is the height of the cylindrical vessel.

For the calculations of the Nusselt number (Nu), if the output of the product Prandtl number (Pr) and Grashof number (Gr) lies between 10⁴ and 10⁹, Eq. 22 will be used, and in case it lies between 10⁹ and 10¹³, then Eq. 23 will be used (Munir et al., 2021): N u 1 = 0.59 ( P r .   G r ) 0.25 . (22) N u 2 = 0.021 ( P r .   G r ) 2 5 . (23)

The convective heat coefficient is calculated as (Munir et al., 2021): h = N u   λ a L . (24)

Using the iterative method, the final value of ‘h’ was calculated until the assumed and the calculated values became equal. The values of convection resistance for walls and top/bottom are calculated with insulation and without insulation for hot water storage tank, yogurt processing unit, the header of ETC, and walls of PVC pipes and tabulated in Table 2.

Radiation losses (Q′ _l-rad ) are calculated by the following relation (Munir et al., 2021): Q l − r a d ′ =   εδA ( T e x t 4 −   T s p a c e 4 ) . (25) where ε is the emissivity, δ is the Stephen–Boltzmann constant, T _space is the space temperature (K), and T _ext is the external area temperature (K).

The external area temperatures are calculated by the following equation (Munir et al., 2021): T e x t = T a m b + ϕ R c o n v . (26)

The values of external area temperatures and radiation losses for walls and top/bottom are calculated with insulation and without insulation for hot water storage tank, yogurt processing unit, the header of ETC, and walls of PVC pipes and tabulated in Table 2.

For the energy analysis, the amount of heat energy transferred and absorbed was calculated. Figure 3 shows the cumulative heat energy released by hot circulating water, heat energy absorbed by milk, and correspondence heat transfer efficiency.

It was found that during the first 40 min of the heating process, hot water released 66, 69, and 74% of its total energy for 50, 40, and 30 L of milk capacities, as shown in Figures 3A,B,C, respectively. The rest of the energy was transferred to milk in the next 100, 70, and 40 min for 50, 40, and 30 L of milk. On the contrary, the milk absorbed 69, 81, and 84% of the total required energy to reach the desired temperature of 80°C. Therefore, the quantity of milk plays an important role in energy consumption. As soon as raw milk was poured into the processing chamber, the rate of energy utilization increased due to the removal of a high quantity of heat. The heat transfer efficiency was found to be above 80% during the whole heating process because milk received energy from the storage tank in which any fluctuation was compensated continuously by ETC. It can be observed that the maximum amount of incoming heat energy absorbed by milk in case of 50 L comparative to the other two qualities, that is, 30 and 40 L, as evident by less gap between the curves of cumulative heat energy absorbed and released. The reason is that the system has been designed for 50 L capacity, and in the case of fully loaded, the heating coil is fully exposed to the milk. Moreover, as soon as the milk temperature reached 70°C, the amount of heat energy absorbed reduced leading to lower heat transfer efficiency. Energy utilization during a hearing process is mainly associated with the temperature difference between circulating hot water and milk as well as milk quality.

The energy used for the heating process followed by fermentation and cooling processes is provided by solar thermal and solar photovoltaic systems. The overall energy consumption for yogurt made from raw milk was calculated to be 6.73 kWh, 5.56 kWh, and 4.21 kWh for 50, 40, and 30 L batches, respectively. Out of this, 40% of energy was consumed during the heating process of 50 L, which was 39% for the cases of 40 and 30 L, while 27% was used for the cooling process of milk (with tap water) for 50 L and 29% for the cases of 40 and 30 L. This shows that the heating process consumed more energy followed by the cooling process in all the cases. During the cooling process of yogurt, 31% of the total energy was consumed for 50 L and 30% for the cases of 40 and 30 L, and the rest of the energy (2%) was used for the fermentation process in all cases. Therefore, the specific production energy (SPE) was calculated to be 485 kJ/kg, 500 kJ/kg, and 505 kJ/kg for 50, 40, and 30 L batches, respectively.

In this study, an evacuated tube collector (ETC) was used for heating applications during the yogurt-making process. A complete algorithm was prepared by taking into consideration useful solar energy available and losses at various sections. The average solar insolation value was taken as 800 Wm⁻² during the experiment. The main important feature of this heating system is the addition of hot water storage which always runs to store hot water in this insulted closed chamber for the effective hot water supply to the yogurt processing unit, but this storage of additional energy has not been taken into consideration. The calculation is based on the step-by-step calculation of available energy and losses at ETC, pipelines, hot water tank, and yogurt processing unit, as shown in Figure 7. In this study, 15 tubes were used, and the total power available on ETC was calculated to be 2218 W. Out of this energy, 5% was the transmission losses, and 5% was the absorbance losses, with a total of 10% of optical losses from ETC. The optical losses were calculated to be 197 W. Water in the tubes was heated due to the thermo-siphon system, and heated water was stored in the insulated top header. The thermal losses inside the header were calculated to be 53 W. So, 1968 W power was available from ETC. The power transferred to the hot water storage tank was 1718 W which was insulated by 50-mm-thick polyurethane. The hot water storage tank stores the thermal energy coming from ETC. The important feature of this hot water storage tank is that it also stores the energy when the yogurt is not processed and always keeps the tank in the thermal potential to be used for processing applications. An electrical coil is also fitted inside the hot water storage tank to heat water in case of cloudy weather conditions. The thermal losses from the hot water tank were calculated to be 127 and 1591 W available for the yogurt processing unit. The line losses in the form of conduction, convection, and radiation in the PVC pipes were found to be 170 W; therefore, the net energy available for the yogurt processing unit was calculated to be 1421 W. Out of this available power for the processing unit, 49 W is thermal losses; therefore, 1372 W is available for useful work. If we calculate the energy required for milk processing/heating, 1350 W power is required to process 50 L of milk/yogurt in 2 h for a temperature differential of 50°C having a specific heat of 3.89 kJ/kg/K. So, the thermal energy available to heat the milk is 1372 W which is greater than the actual power demand, thus justifying the number of tubes and the size of ETC. We may assume this additional energy as undetermined losses, and this power is utilized for thermal storage in the hot water tank. The complete distribution is shown in Figures 7A,B.

Table 3 summarizes some of the related studies regarding thermal analysis of yogurt/milk processing for a comparative assessment of the current solar yogurt processing unit. Although the quantity of the product to be processed, design of the process, and operating conditions vary and are important for the rates of energy usage, it can be observed that the solar yogurt processing unit gave quite good results. The exergetic efficiency of the current unit is higher than all the tabulated other studies. Even dealing higher quality of product, the specific exergy consumption is less than that of the processing plant reported by Mojarab et al. (2016) while almost similar to that reported by Mojarab & Aghbashlo (2017). Compared with the industrial-scale yogurt processing plant, the overall exergy efficiency and specific exergy consumption are less. Although the difference between the values of specific exergy consumption is almost 50% in comparison to the large difference of quantity being dealt into both cases; however, it is understandable that processing large quantities leads to decreased specific energy consumption. Abu-hamdeh (2020) conducted an exergetic analysis on the combined cooling, heating, and power (CCHP) system based on a PVT collector. It can be observed that the specific energy consumption rate of hot water was 3.78 kW for 350 L of water capacity, while in the current study the specific energy consumption rate of milk was 2.89 kW for 50 L batch capacity. Exergy extracted by hot water was 0.2343 kW which lies in the range of the current study (0.019–0.374 kW for 50 L), and exergy extracted during refrigeration was 0.09122 kW which is lower than that reported in the current study (1.76–1.08 kW for 50 L). It was because of the use of a vapor compression refrigeration system (COP 4.33) in the current study, while a vapor absorption refrigeration system (lower COP of 0.615) was used by Abu-hamdeh (2020).