Investigating the impact of using nano-fluid as a cooling medium on photovoltaic/thermal panel system performance

Making use of the superior thermal properties of nanofluids is now very common, especially with regard to the cooling of photovoltaic panels to improve overall efficiency. In this work, a novel cooling system manufactured from 3 mm aluminium was attached to the rear of a monocrystalline photovoltaic module, and two volumetric concentrations of SiC/Water nanofluid were tested with different flow rates. These tests were carried out under outdoor climatic conditions in middle of Iraq at Babylon University (32.46 °N, 44.42°E) during both winter and summer. A theoretical model was thus developed in SolidWorks and simulated using ANSYS 18.2. The maximum enhancements in electrical and overall efficiency were found to be 50% and 82.41% in March with a 0.5% nanofluid concentration and 2 L/min flow rate, while the minimum enhancements were 35.4% and 34.01% in June, with a 0.1% nanofluid concentration and 0.5 L/min flow rate. The theoretical results showed good approximation to the experimental results, and the average deviation percentage of electrical efficiency for a photovoltaic/thermal system with nanofluid on 27 March was 5.58%, while on 3 June it was 11%.


Introduction
The use of conventional energy resources is the main reason for global warming. To slow climate change, implementing renewable energy is thus vital. Utilising photovoltaic modules is one way of accessing clean energy resources used globally, as photovoltaic cells absorb incident radiation from the sun and convert it to electrical and thermal energy. Part of this absorbed radiation is converted to electricity, while the remaining percentage is converted to heat inside the cells. According to the simple diode equation which is the basis of photovoltaic modules, such increases in cell temperature decrease the open circuit voltage, thus reducing the electrical efficiency, however. Photovoltaic/Thermal (PV/T) solar systems thus generally incorporate cooling systems to cool the modules and increase their electrical efficiency. Recently nanofluids have become widely used as coolants due to their high thermal conductivity, which enhances the heat transfer. Abd-Allah et al. [1] performed research at Benha University, Egypt with the aim of investigating the impact of cooling PV cells with Boehmite-water nanofluid (AlOOH-xH2O) as a coolant. A three-dimensional model of the PV was designed in ANSYS-FLUENT with given boundary conditions and single phase flow of 0.08, 0.1, 0.2, 0.3, and 0.4 L/min in straight channels by solving the governing equations of energy conservation, mass conservation, and IOP Publishing doi: 10.1088/1757-899X/1067/1/012118 3 attached to a sheet-and-tube cooling panel and circulating pumps. The average increases in electrical efficiency were 5.48%, 6.54%, 6.46%, and 6.36% for deionized water, TiO2, ZnO, and Al2O3, respectively. Younis et al. [7] studied the influence of using Al2O3-ZnO-H2O Nanofluid on the thermal and electrical performance of a PV/T system. Experiments were achieved on a setup consisting of an outdoor set of five solar PV/T hybrid collectors connected in series, with closed hydronic loops in which the flow rate was constant at 3.25 L/min, to an indoor storage tank with an integrated heat exchanger. The mass fraction of Al2O3 was 0.05% with a particle size of 5nm and the mass fraction for the ZnO was 0.05%, with a particle size of 10 to 30 nm. The results revealed that the average increase in the electrical efficiency was less than 0.1% in comparison with a module without cooling, which was considered to be due to the low concentration of nanoparticles. The average increase in thermal efficiency was 4.1%.Nasrin et al. [8] performed an indoor study to improve the thermal performance of PV/T systems using a water/multi wall carbon nano-tube nanofluid as a coolant. The system consisted of a PV panel and a thermal collector manufactured from aluminium tubing (serpentine) with no absorber plate. The nanofluid was cooled using a cross flow air heat exchanger with fan, and the radiation simulation system consisted of 120 halogen bulbs to provide light with different intensities. The results revealed that the thermal and electrical efficiencies were increased by 0.14 % and 3.67%, respectively, for PV/T systems with water/MWCNT nanofluid (1% wt.) in comparison with those for pure water. Shahad et al. [9] performed an experimental study at Babylon University, Iraq, to investigate the impact of water cooling on the performance of PV/T systems under Iraqi weather conditions. They used an acrylic pocket type collector fixed to the back of the PV panel so that there was direct contact between the tedlar layer and the cooling water. Different flow rates were tested (0.5, 1, 1.5, and 2L/min), with results that revealed that the maximum enhancements in electrical and overall efficiencies were 18.68% and 81%, respectively in March at a flow rate of 2 L/min. The minimum enhancements in electrical and overall efficiencies were 13.36% and 74.08%, respectively, in July at a flow rate of 0.5 L/min. Michael and Iniyan [10] performed a study in Anna University, India that analysed a PV/T system by fixing a copper thermal collector directly to the photovoltaic cells and removing the tedlar layer, thereby reducing the thermal resistance. They used CuO/water Nanofluid as a coolant with a constant mass flow rate of 0.01 kg/s, and the volumetric concentration of nanoparticles was 0. 05%. NaOH was used as a stabilising agent, added slowly during the preparation of the nanofluid until the pH of the solution was between 6 and 7. The collector consisted of a one water channel with depth of 0.002 m, which was manufactured from aluminium and fixed to the back of the PV. The results revealed that the nanofluid increased the thermal efficiency of the PV/T system to 45.76%. The electrical efficiency of the PV/T system was reduced, however, due to the high temperature of the inlet water, prompting a recommendation to use an effective heat exchanger.
The objective of the current study is to investigate the impact on the performance of photovoltaic/thermal solar system of using a SiC/water nanofluid as a coolant.

Experimental Setup
A rig was manufactured and installed in Babylon University campus, Iraq (32.46 °N, 44.42°E), consisting of three identical monocrystalline photovoltaic modules. Two of these were modified to have a pocket aluminium collector, with one cooled using SiC/Water nanofluid and the other by pure water, while the back sheet of the third PV module was cooled by the surrounding air, as shown in figure 1. The nanofluid was circulated by pump and cooled by a helical heat exchanger, as shown in figure 2. The tilt angle was adjusted monthly according to the inclination angle, while the PV modules were directed to the south (zero azimuth angle). Figures 2 and 3 show the schematics of the rig.    Figure 3. Schematic of the PV/T system

Nanofluid Properties
The fluid investigated in this research was SiC/water nanofluid. Dispersing Silicon Carbide nanoparticle in pure water enhances its thermal conductivity and improves heat transfer. The thermal properties of the pure water at a temperature of 298 K [11] are shown in table 1, while the thermal properties of SiC nanoparticles [11] are shown in table 2. Two volumetric concentrations were used in this research, 0.1% and 0.5%. The properties of the nanofluid were calculated using the following equations: 1. Volume fraction [12] ( )

Specific heat [12]
( The thermal properties for the two volumetric concentrations are thus shown in table 3 Table 3. Properties of SiC/water Nanofluid

Model Geometry
Drawing the model geometry is the first step in the simulation process. Figure 4 shows a sketch of the model geometry, with suggested PV/T and PV/TN models that include a glass cover, photovoltaic cells, a tedlar layer, an aluminium cooling panel, and cooling fluid. The geometry and the domain of the model were designed by the SolidWorks 2016 and then imported to an ANSYS 18.2 workbench. The characteristics and dimensions of various layers of the model are shown in Table 4, while the model geometry and three model domains are presented in figure 5. The properties of the solid layers are presented in table 5.   Figure 5. Geometry and Domains of Three PV Systems

Simulation Assumptions
1. The fluid is incompressible, with uniform properties, and the system is in a quasi-steady state 2. The pure water and nanoparticles are in thermal equilibrium, forming a single phase nanofluid.
3. Flow at the inlet is one-directional. 4. There is a laminar flow region. 5. Some part of the energy entering PV cells is converted into electrical energy, with the remaining portion contributing to an increase in the temperature of cells and nanofluid. 6. Solar irradiance is perpendicular to the PV panels. 7. Radiation heat loss from PV panel is negligible. 8. The bottom surface of the PV/T and the sides of the cooling panels are adiabatic.

Governing Equations
The governing differential equations as solved by ANSYS are the continuity equation, energy equation, and momentum equation. The pressure-velocity based solver was selected, which solves the momentum equation to obtain the velocity field. For the pressure field, this was extracted by solving the pressure correction equation obtained by manipulating the continuity and momentum equations. The energy equation allowed the setting of input parameters related to heat transfer in the boundary conditions and for outputs related to energy such as temperature to be obtained, as shown in Figure 6. The first layer is glass, which absorbs the heat from solar radiation; the transfer of the heat in this layer is by conduction, as shown in equation (6) [13]:

2-PV layer
At this layer, heat transfer is by conduction, as represented in equation 7:

3-Tedlar layer
In this solid domain, the energy equation is

4-Aluminium plate (Al)
The energy equation in this domain is The momentum equations in the three Cartesian coordinates are thus • -direction: • -direction: • -direction: The continuity equation in Cartesian coordinates is The effects of surface stresses are accounted for explicitly, while the source terms SMx, SMy, and SMz in equations 11, 12, and 13 include contributions from body forces only. For example, the body force due to gravity is modelled by SMx = 0, SMy = -ρ g cos β, and SMz = -ρ g sin β.
The gravity force can significantly influence the buoyancy force and, consequently, the convection terms. Thus, gravitational acceleration, which is a function of the setup tilt angle, must be activated in the simulation procedure. The vectors of gravity acceleration are defined as [2] .sin

Grid Study
A grid independence test was implemented to ensure the accuracy of the calculated solutions. The simulation results for six different mesh sizes for average back sheet temperature were utilised, and the results of the mesh refinement study are shown in table 6. Further refinement on the grid with 8,104 nodes showed no considerable effect on the numerical solutions, merely increasing the computational cost and time. The mesh with 8,104 node elements was thus selected as an appropriate computational domain, as shown in figure 7.   Characteristics of PV cells affected by irradiance and temperature were modelled using a circuit model. The PV module has a non-linear voltage-current (V-I) characteristic, modelled using current sources, diodes, and resistors. A single-diode model was used to simulate PV characteristics in the PVsys software.
The following I-V relationships were used to predict the electrical power output of the PV equivalent circuit [14]:    Figure 8. Definition of a PV module in PVsys

4.Results and Discussion
The experimental tests were performed during selected days in February, March, and June. The same climatic conditions were thus applied to CFD simulation to allow comparison with the experimental results. The high thermal conductivity of the SiC/water nanofluid improved the heat transfer rate. Figure 9a shows that, in the non-cooled PV module, the temperature field was symmetrical about the axes of the module due to similar boundary conditions all around causing a temperature gradient such that the maximum temperature was in middle of PV module. Figures 9b and 9c show the lowest temperature at the inlet of the cooling fluids which absorb the heat from module; thus, the temperature increases in the direction of flow in these cases.  It was noted experimentally that the back sheet temperature of the PV modules was influnced by many factors, including ambient temperature, solar radiation, the temperature of the inlet cooling fluid, and its flow rate. High solar radiation increased the temperature of the back sheet due to larger part of the nonconverted solar energy being stored as heat inside PV cells. Low ambient temperature increased the rate of heat transfer from the PV module by convection. The wind speed in Iraq has no major influence, however, due to thehigh ambient temperature, especially during summer. Figures 10 and 11 show the variations in the measured average back sheet temperatures with flow rates for PV/TN and PV/T systems, respectively. The average back sheet temperature of the PV/TN system was lower than that of the PV/T system for all flow rates and nanofluid concentrations due to the high thermal conductivity of SiC/water nanofluid, which improves the rate of heat transfer. Increasing the flow rate of both pure water and SiC/water nanofluid decreases the back sheet temperature due to forced convection, however.  Efficiency   Figures 12, 13, and 14 show the experimental variation of measured instantanous electrical efficiency during selected days in March and June for PV, PV/T, and PV/TN systems, respectively. Electrical efficiency was high in the morning due to lower ambient temperatures, while the electrical efficiency during March was higher than that during June because the solar radiation in March is relatively higher than that in June. The ambient temperature in March is also lower, affecting heat transfer by convection and the temperature of the inlet cooling fluid for PV/T and PV/TN systems. The electrical efficiency of the PV/TN system was higher than that of the PV/T system due to the high thermal conductivity of the SiC/water nanofluid, which increases the heat transfer rate and cools the back sheet. Using a 0.5% concentration of SiC/water nanofluid in March has clear impact on enhancing heat transfer as compared with the 0.1% concentration. Good agreement was found between the predicted and measured results for the PV module, with an average deviation of 8.9% in March and 13.26% June. For the PV/T system, the average deviation in March was 6.63, with 9.21% for June, and for the PV/TN system the average deviation in March was 5.58%, with 11% in June.   Figure 15) shows the experimental daily average electrical efficiencies for the PV, PV/T, and PV/TN systems in two months. The maximum daily average efficiencies were 7.31%, 9.4%, and 10.96% respectively in March, due to low ambient temperature which improves energy conversion, while the minimum daily average efficiencies were 6.33%, 8.24%, and 8.58%, respectively in June due to high ambient temperatures, which reduce heat transfer by convection and increase the temperature of inlet cooling fluids  Figure 16 shows the daily average thermal efficiencies of PV/T and PV/TN systems for selected days in March and June with 0.1% volumetric concentrations of SiC/water nanofluid. Various volumetric flow rates of pure water and nanofluid were tested (0.5, 1, 1.5, and 2 L/min). The daily average thermal efficiency of the PV/TN system wass higher than that of the PV/T system for all flow rates due to the high thermal conductivity that improves heat gain and increases the nanofluid outlet temperature. In March, with flow rates of 0.5, 1, 1.5, and 2 L/min, the daily average thermal efficiencies of PV/T and PV/TN were 40.67%, 50.67%, 58.52, and 63.57% and 48.91%, 54.07%, 60.89%, and 68.52%, respectively, while in June with the same flow rates, the daily average thermal efficiencies of PV/T and

1)
The maximum temperatures of the PV panel were 63.5 °C and 75.2 °C for March and June, respectively.

2)
The maximum temperatures of PV/T system were 39.1 °C and 45.6 °C for March and June, respectively.

3)
The maximum temperatures of the PV/TN system were 37.9 °C and 44.3 °C for March and June, respectively.

4)
The average electrical power of the PV/TN system was higher than that of the PV/T system because the SiC/water nanofluid has higher thermal efficiency, which increases the heat transfer rate and decreases the back cell temperature.

5)
The maximum enhancements in electrical efficiencies for PV/TN and PV/T were 36.7% and 31.3%, respectively, in March, while the minimum enhancements were 35.4% and 30.1%, in June.

6)
The maximum enhancements in overall efficiencies of PV/TN and PV/T were 82.41 % and 73.77 %, respectively, in March, while the minimum enhancements were 34.01% and 31.66%, in June.