## Performance analysis of a novel integrated photovoltaicâ€“thermal system by top-surface forced circulation of water

**Abstract**

Abstract Almost 80–90% of energy is wasted as heat (provides no value) in a photovoltaic (PV) panel. An integrated photovoltaic–thermal (PVT) system can utilize this energy and produce electricity simultaneously. In this research, through energy and exergy analysis, a novel design and methodology of a PVT system are studied and validated. Unlike the common methods, here the collector is located outside the PV panel and connected with pipes. Water passes over the top of the panel and then is forced to the collector by a pump. The effects of different water-mass flow rates on the PV panel and collector, individual and overall efficiency, mass loss, exergetic efficiency are examined experimentally. Results show that the overall efficiency of the system is around five times higher than the individual PV-panel efficiency. The forced circulation of water dropped the panel temperature and increased the panel efficiency by 0.8–1% and exergy by 0.6–1%, where the overall energy efficiency was ~81%. Graphical Abstract Open in new tabDownload slide integrated photovoltaic–thermal, forced cooling, energy analysis, exergy analysis, flat-plate collector Introduction Bangladesh is confronting a great deal of energy emergencies and genuine desertification issues in provincial areas. These issues could be ameliorated if sustainable power sources are utilized as an essential source of energy in rural regions. Although Bangladesh has a considerable amount of fossil resources, the amount is degrading to a great extent as the dependency on it is remarkable. For instance, the primary sources of energy in this country are natural gas (60%) followed by hydropower and coal, which are probably going to be exhausted very soon due to their extensive use [1]. Therefore, if no advanced innovation is introduced, then Bangladesh will face a tremendous energy crisis in the future. In these cases, sustainable power sources are the only hope for the general population of Bangladesh. Individuals have an expansive unsatisfied need for energy that is developing by 10% yearly [2, 3]. In the last few years, the government has taken several initiatives to address the energy crisis. Not only in the public sectors, but also this issue is given much importance from different individual sectors. Although power generation in the most recent years has increased a lot, still it is not enough to face the soaring demand of the country. Moreover, Bangladesh has the lowest per capita consumption of energy in South Asia [3]. Presently, the total generation capacity is 15 821 MW [1, 4]. Coal, gas and diesel are being used in Bangladesh for producing electricity as primary resources. At present, there is a huge gap between production and demand. The demand is increasing day by day and there is a prediction that it will reach ~40 000 MW by the year 2030 [5]. In this circumstance, different public and private associations must work either together or individually to overcome this problem [6, 7]. However, it could be promising that Bangladesh is a semi-tropical region lying in north-eastern South Asia. The normal daylight duration in Bangladesh during the dry season is ~7.6 hours and in the rainy season is ~4.7 hours. The most noteworthy daylight hours are received in Khulna, with readings extending from 2.86 to 9.04 hours, and in Barisal, with readings going from 2.65 to 8.75 hours (wet to dry seasons). These amounts can be compared with the 8 hours of daily sunshine in Spain, which allowed the generation of 4 GW of solar power (2.7% of the national capacity) before the end of 2010 [7]. The other parts of Bangladesh also get a significant amount of solar energy. However, in the rainy season, the amount reduces but still it is satisfactory to implement different solar-powered devices for producing electricity. The photovoltaic (PV) panel is a promising device for producing electricity from solar radiation. The current efficiency range of a usual PV panel remains within 10–20% [8]. Thus, almost 80–90% of the incident energy is lost in a PV panel. This waste energy can be recovered using an integrated photovoltaic–thermal (PVT) system. Several pieces of work have been done on this type of system in the last couple of years. Ibrahim [9] experimented on a glazed and unglazed photovoltaic–thermal water-heating (PVT/W) system along with enhancing the conductivity by using an aluminium reflector. It is worth mentioning that PVT/W refers to a system that includes a hybrid PVT system with the thermal unit of water circulation through a heat exchanger. The study opined that using aluminium can be more efficient than using other reflectors. A similar type of experiment was conducted by Mojumder et al. [10] with a solar simulator. Fudholi et al. [11] conducted an experiment using a water-cooling system in the solar-radiation range of 500–800 W/m2 and the efficiency was found to be 68%. Khanjari et al. [12] studied thermal-efficiency enhancement by using nano fluid with water in a PVT system. The study showed that thermal efficiency can be increased by increasing the volumetric ratio of the nano particle. Al-Waeli et al. [13] did an intensive review that discussed the previous methods along with their implementation techniques and efficiencies. The study explained that the thermal efficiency is increased by using nano fluid in the system and the unglazed PVT system provides more energy than any other system. Besides those, many other different methods are used to improve heat transfer. Some more recent work can be found in the literature [14–21]. Most of the studies discussed above were used for domestic purposes. From an energetic point of view, these systems are more effective than using conventional solar thermal collectors and PV components. But it is a matter of fact that the collector has been positioned below the PV panel from the beginning of using the PVT system. In this kind of system, a collector is connected to the PV system. Water is supplied from the back of the PV panel, extracting heat and cooling it down. However, the efficiency, installation and economical condition are important issues in this type of conventional system. The solar collector positioned below the PV panel cannot get solar radiation directly. As a result, the efficiency of the collector is less than the standard efficiency (60–70%). Moreover, the lost heat of the PV module is absorbed by the collector. Hence, the efficiency of the PV panel can be increased but the collector exhibits comparatively lower efficiency than the standard one. In addition, extra heat-transfer materials need to be used for improving the heat transfer from the PV panel to the collector. From the recent work of Wu et al. [22], a new method of incorporating the technique of flowing the water through the top surface of the PV panel and using a water-cooled type of collector can play a significant role in improving the efficiency of a PVT system. The study showed the effect of the variation of solar radiation and the height of the cooling channel on the characteristics of the heat transfer. The collector used in this study is placed outside the panel; therefore, the solar radiation can incident directly on the collector. Hence, the efficiency of the collector seemed to be increased, which also results in the increasing of the overall efficiency. However, the proposed system has yet to be developed satisfactorily, as the study only showed the numerical analysis. Rigorous analysis considering the efficiency of the PV panel and collector; the mass flow rate; the loss of energy due to evaporation, radiation and convection; and the distribution of the temperature of the water over the panel and inside the collector need to be made using a developed experimental set-up. In 2019, Arefin [23] conducted experiments on a top-surface water-cooling method for validating the system. However, the author used the natural circulation of water; therefore, the mass flow rate of the water was lower than in a forced circulation system. In the proposed system of this work, the collector is located outside the PV panel and better integrated with the panel as compared to other systems. First, the water passes over the panel, extracts heat from the panel and cools it down. Then the water passes through the collector and hot water is produced. A pump is used for maintaining the force circulation of the water. In this research, mathematical and experimental investigations are performed on an integrated PVT system by the top/front-surface forced circulation of water from energy and exergy perspectives. The mass flow rates of 2 l/min (9.09 × 10–4 l/min/cm2), 2.25 l/min (1.02 × 10–3 l/min/cm2), 2.5 l/min (1.14 × 10–3 l/min/cm2), 2.75 l/min (1.25 × 10–3 l/min/cm2) and 3 l/min (1.36 × 10–3 l/min/cm2) are maintained. The temperatures of the panel, heat loss, panel efficiency, collector efficiency, mass loss, overall system efficiency, electrical exergetic efficiency and the share of energy and exergy loss are evaluated. For accuracy, experimental data are taken for several months. This is the very first experimental research conducted to validate this type of system with the forced circulation of water. 1 Mathematical modelling In this section, the mathematical modelling of the entire system is discussed from the viewpoint of energy and exergy analysis through theoretical and empirical equations. Every subsystem is evaluated and a model of the entire system is presented in this section. Mathematical modelling is very much necessary in carrying out the further work of this study. 1.1 Photovoltaic panel-heating-rate model To determine the cooling frequency of the PV panel, the heating rate of the panel is used. The heating rate of the PV module can be specified by calculating the module temperature as a function of time. Equation (1) can be used to determine the module temperature [24–27]: Tm=Tamb+(NOCT−25)E/80(1) From Equation (1), it is clear that the module temperature depends on the solar irradiance, ambient temperature and normal operating cell temperature (NOCT). The NOCT is the function of the ambient temperature at the sunrise time. NOCT=25oC+Trise(2) The value of NOCT is constant but ambient temperature and solar radiance are variable. 1.2 Modelling for PV-module surface cooling and energy performance The PV system extracts energy directly from solar radiation. Some of the energy is converted into electrical energy and most is converted into heat. The water gains heat by two means: (i) direct solar radiation and (ii) extracted heat from the PV panel. A very small amount of water gets evaporated due to heat. There are several assumptions for this model: Loss due to the viscosity of the water, pipe friction and leakage is ignored. The film thickness of the water over the entire length of the panel is entirely uniform. Reflection due to the front glass is assumed to be 10% [28, 29]. The ohmic loss is not considered, as it is very low [30]. The system is considered to be in a quasi-steady state. The energy balance is shown in Fig. 1. Fig. 1: Open in new tabDownload slide Energy balance for the water in the PV module The water gains energy from sunlight directly and from the front glass. The reasons for the lost energy have been already stated. So, the energy balance for the water and the front glass is: mwcpwdTdt=(haw−Qrad+Qcom−gw−Qconv.−water−Qevaporation)Ac+(mincpwTs)−(moutcpwTw)(3) The solar radiation absorbed by the water directly per unit area is: haw=αahbt+αdhdt(4) The surface of the PV panel is taken as inclined. When the PV panel is positioned horizontally, it gains solar radiation properly when the Sun is at the top of the head. For this type of surface, the diffuse-radiation equation and behaviour were proposed by Ma and Iqbal [31]. The formula for the hourly diffuse-radiation incidence is: Is=Id(1+cosθ)2(5) and Hs=Hd(1+cosθ)2(6) In this model, the intensity of the diffuse radiation is considered to be independent of the azimuth and zenith angles. In the case of a partly cloudy sky, Krauter’s model [28] can be used to predict the diffuse radiation. Krauter’s formulation for an inclined surface is: Is=Id[(1+cosθ)/2](1+Fsin5(θ/2)]×[1+Fcos2θsin3θ](7) From calculation, it is found that the Reynolds numbers for stream velocities of 2 l/min (9.09 × 10–4 l/min/cm2), 2.25 l/min (1.02 × 10–3 l/min/cm2), 2.5 l/min (1.14 × 10–3 l/min/cm2), 2.75 l/min (1.25 × 10–3 l/min/cm2) and 3 l/min (1.36 × 10–3 l/min/cm2) are 1 425 022, 1 603 148, 1 781 276, 1 959 403 and 2 137 531, respectively, which implies that the flow is turbulent. Here, the dynamic viscosity of the water was taken at 28ºC. Therefore, for the entire range of Reynolds numbers, the Nusselt number is: Nu= {.825 +0.387 RaL16{1+(0.492pr)916}827}2(8) pr is the Prandtl number. The Penman–Monteith equation describes evaporation from a water surface. The equation depends on air pressure, temperature, wind speed and solar radiation. It is widely regarded as one of the most accurate modes in terms of estimates [32, 33]. A simpler equation to estimate the evaporation between water and ambient air is: Qevp.=26.639×101−V0.05(ρw−ρd)×hfgpT(9) The energy balance from the front glass is: mgCpgdTgdt =(Qcon.cg−Qconv.pw) Aactual(10) Aactual is the actual effective area (m2). Water is stored in a tank. There is an energy term for the water inlet and for the outlet. The evaporation loss from the tank depends on the insulation. For a highly insulated tank, this loss is very small. For evaporative loss, Equation (9) can be used. PV-module electrical efficiency is calculated by using Equation (11). ηelectrical=I ×VIs ×Aactual(11) 1.3 Modelling for a flat-plate collector and overall system efficiency The solar radiation received by a flat-plate collector is: QC=I×A(12) Part of the radiation is absorbed by the glazing, which also increases the temperature, part is reflected back and the rest is transmitted through the glazing. The percentage of solar-ray penetration and the percentage of the absorption of rays are indicated by the conversion factor. It is the product of the absorption rate of the absorber and the rate of transmission of the cover: Qc=I×(τα)×A(13) Some portion of the heat is lost by convection and radiation. The rate of heat loss (Qo) mainly depends on the overall heat-transfer coefficient (UL) and the collector temperature: Qo=ULA(Tc−Ta)(14) The rate at which useful energy (Qu) is extracted by the collector is expressed as a rate of extraction under the steady-state condition, which is proportional to the rate at which useful energy is absorbed by the collector + the amount of energy lost to the surroundings: Qu=Qi−Qo=I×(τα). A−ULA(Tc−Ta)(15) The rate of extraction of heat from the collector is indicated by the amount of heat carried away by the fluid passing through it: Qu=mcp(To−Ti)(16) It is very difficult to determine the average collector temperature in some cases. Therefore, it is necessary to relate the useful energy gain of a collector to the overall useful energy gain if the whole collector surface were at the fluid-inlet temperature. Thus, a new term has been defined, called the ‘Collector heat removal factor (FR)’: FR=mcp(To−Ti)A[Iτα−UL(Ti−Ta)](17) The actual useful energy of the collector can be obtained by Equation (18). This is a widely used method to determine the collector energy gain and is known as the ‘Hottel–Whillier–Bliss equation’ [34]: Qu= FRA[Iτα−UL(Ti−Ta)(18) The instantaneous overall thermal efficiency of a collector is: ηcollector=FRA[Iτα−UL(Ti−Ta) AI(19) ηcollector=FRτα−FRUL{(Ti−Ta)I}(20) The overall system performance is defined by Equation (21), which is the summation of the electrical efficiency of the panel and the thermal efficiency of the collector. Electrical efficiency is given in Equation (11). ηTotal=ηcollector+ηelectrical(21) 1.4 Exergy analysis Based on Fig. 2 and considering a quasi-steady-state condition [35], the general equation for the exergy analysis of a PVT system can be defined as: Fig. 2: Open in new tabDownload slide Schematic diagram of exergy-balance analysis ∑E˙xin= ∑E˙xout+∑E˙xloss(22) E˙xsun+E˙xmass,in=E˙xel+E˙xmass, out+∑E˙xdest(23) ∑E˙xdest=E˙xdest,radiation+E˙xdest,convextion +E˙xdest, friction (24) where E˙xin denotes the input energy rate, E˙xout is the output energy rate and ∑E˙xdest is the destroyed energy rate. Also, E˙xdest,radiation , E˙xdest,convextion and E˙xdest,friction indicate the radiation, convection and friction losses, respectively. Here, the input exergy rate extracted from the Sun can be defined as [36]: E˙xsun=G˙(1 − T˙ambTsun)(25) Here, Tamb and Tsun are the temperatures of the surroundings and the Sun, respectively. In addition, the exergy of the mass flow rate is defined as follows [36, 37]: E˙xmass, out−E˙xmass, in=m˙f(φout−φin)(26) where φout=(hout−hamb)−Tamb(Sout−Samb)(27) φin=(hin−hamb)−Tamb(Sin−Samb)(28) where h and S are defined as entropy and specific enthalpy, respectively. The electrical exergy of a PVT is defined as [38]: E˙xel=E˙el(29) The overall exergy of a PVT system can be obtained as follows: E˙xov=E˙xel+E˙xth(30) The destroyed energy given in Equation (31) can be obtained by using the following equation: ∑E˙xdest=E˙xsun−E˙xel−E˙xth(31) Here, the electrical ( εel ), thermal ( εth ) and overall exergy ( εov ) efficiencies of PVTs can be obtained by the following equations [39]: εel=E˙xelE˙xsun=E˙xelG˙(1−T˙ambTsun)=Voc × Isc × FFG˙(1−T˙ambTsun)(32) εth=E˙xthE˙xsun=m˙f.Cpf[(Tf,out − Tf, in) − Tambln(Tf,outTf, in)]G˙(1 − TambTsun)(33) εov=εel+εth(34) 1.5 Uncertainty analysis For ensuring the reliability of the experiments, an uncertainty analysis is conducted on thermal and electrical efficiencies. If K is a function of ‘n’ independent linear parameters such that K = K (s1,s2,…..sn), the uncertainty of function K is defined as: δ K= (∂K∂s1 δs1)2+ (∂K∂s2δs2)2+…+ (∂K∂snδsn)2(35) where ∂K is the uncertainty of function K and ∂Vi is the uncertainty of Vi. whereas, ∂K∂si is the partial derivative of K with respect to si. More details about uncertainty analysis can be found in the literature [40]. On the basis of the uncertainty analysis, it is found that the maximum absolute uncertainties for all parameters are <2.5%. 2 Materials and method The main parts of the experimental set-up are (i) a flat-plate collector, (ii) a PV panel, (iii) a water tank and (iv) a storage tank. All parts of the experiment are linked through a piping network. The water tank (80 l) is situated ~1 m above the panel. In this experiment, water from the tank goes to the PV module with the help of pipes. The water flows down over the top/front surface of the PV module. For this purpose, a PVC pipe with 20 holes is placed on the top of the PV module to maintain a constant discharge of water. The water cannot flow out from the side of the PV panel because of a tray placed at the bottom side of the module that also provides support to the module. Tables 1 and 2 show the collector and PV-panel properties. First, water goes over the PV panel and extracts heat from the panel. Then, water is forced through the pump to the collector and is heated up in the pipes of the collector. Finally, the heated water is collected from the collector outlet into the storage tank. A pump is used to force the water to the collector at a different mass flow rate. Fig. 3 shows the schematic and the experimental set-up. Table 1: Solar PV-panel specifications Parameters . Dimensions . Solar panel 1 piece Rated output 20 W (total) Dimensions 50 × 44 × 5 cm No. of cells 108 Efficiency Variable Battery 12 volts, one piece Parameters . Dimensions . Solar panel 1 piece Rated output 20 W (total) Dimensions 50 × 44 × 5 cm No. of cells 108 Efficiency Variable Battery 12 volts, one piece Open in new tab Table 1: Solar PV-panel specifications Parameters . Dimensions . Solar panel 1 piece Rated output 20 W (total) Dimensions 50 × 44 × 5 cm No. of cells 108 Efficiency Variable Battery 12 volts, one piece Parameters . Dimensions . Solar panel 1 piece Rated output 20 W (total) Dimensions 50 × 44 × 5 cm No. of cells 108 Efficiency Variable Battery 12 volts, one piece Open in new tab Table 2: Specifications of flat-plate collector Parameters . Dimensions . Width 68 cm Length 102 cm Tube diameter 0.635 cm Inlet and outlet diameter 0.635 cm Total area 8000 cm2 Effective area 6936 cm2 Parameters . Dimensions . Width 68 cm Length 102 cm Tube diameter 0.635 cm Inlet and outlet diameter 0.635 cm Total area 8000 cm2 Effective area 6936 cm2 Open in new tab Table 2: Specifications of flat-plate collector Parameters . Dimensions . Width 68 cm Length 102 cm Tube diameter 0.635 cm Inlet and outlet diameter 0.635 cm Total area 8000 cm2 Effective area 6936 cm2 Parameters . Dimensions . Width 68 cm Length 102 cm Tube diameter 0.635 cm Inlet and outlet diameter 0.635 cm Total area 8000 cm2 Effective area 6936 cm2 Open in new tab Fig. 3: Open in new tabDownload slide (a) Schematic representation of the experimental set-up and (b) the experimental set-up The primary objectives of this study are to increase the photovoltaic panel efficiency by forced water circulation as well as utilizing the heat loss from the collector to produce hot water. The efficiency and temperature ranges of the panel were determined for various tilt angles to find the optimum tilt angle suitable for Rajshahi (Bangladesh). The water was supplied at various flow rates: 2 l/min (9.09 × 10–4 l/min/cm2), 2.25 l/min (1.02 × 10–3 l/min/cm2), 2.5 l/min (1.14 × 10–3 l/min/cm2), 2.75 l/min (1.25 × 10–3 l/min/cm2) and 3 l/min (1.36 × 10–3 l/min/cm2) for 2 min at 15-min intervals considering the fact that continuous flow will reflect some portion of the solar radiation, which would adversely affect the efficiency of the panel. The time interval for the valve opening was maintained by importing a simple time-delay code embedded in the automatic microcontroller system called Arduino. In the final step, a thermometer with an accuracy of ±1°C was used to measure the temperature of the hot water from the collector that accumulated in a storage tank. For accurate results, the readings were carefully taken for several months from 10 AM to 4 PM as well as a pyrometer (PMA 2144 class) device (accuracy ±10 W/m2) was used to measure the effective solar intensity. 3 Results and discussion Table 1 (in the online Supplementary Data) exhibits sample readings of the experiment. Five plots (Fig. 4) show how the panel temperature differs at different times during the day and the impact of water cooling on the panel temperature. It is apparent that, as the mass flow rate of the water stream increases, the panel temperature drops. With the increase in the mass flow rate, more water can extract more heat from the PV panel and eventually reduce the temperature. This is because the water has the highest specific heat and absorbs heat while flowing across the panel. By increasing the flow rate of the water, the larger amount of water absorbs more heat form the panel. For a mass flow rate of 2.5 L/min (1.14 × 10–3 l/min/cm2), at best, a 15°C panel temperature can be reduced, which significantly improves the efficiency of the PV panel. The panel temperature directly affects the efficiency, as seen in Fig. 5. The lower the panel temperature, the higher the efficiency. However, the efficiency also depends on the radiation of the sunlight falling on the unit area of the panel. So, efficiency will be at the highest value when the panel and the collector get the highest radiation from the Sun. The radiation of sunlight falls at its highest at noon each day. It is also evident from the figure that, for a reduction of 1°C in the panel temperature, the efficiency increases by 0.3%. Considering the relationship between efficiency and temperature, efficiencies of 14.6% and 15.2% are obtained for respective panel temperatures of 48.3°C and 40°C in this experiment. Fig. 4: Open in new tabDownload slide Panel temperature at (a) mw = 2 L/min (9.09 × 10–4 l/min/cm2), (b) mw = 2.25 L/min (1.02 × 10–3 l/min/cm2), (c) mw = 2.5 L/min (1.14 × 10–3 l/min/cm2), (d) mw = 2.75 L/min (1.25 × 10–3 l/min/cm2) and (e) mw = 3 L/min (1.36 × 10–3 l/min/cm2) Fig. 5: Open in new tabDownload slide Relation between PV-panel efficiency and temperature Fig. 6 shows the efficiency of the panel with and without water cooling along with the variation in tilt angles. However, the relation between the tilt angle and the efficiency is not linear. Up to an angle of 45°, the panel efficiency increases and, above 45°, the efficiency decreases gradually. This is because the panel will work at the optimum condition when it gets the solar radiation that is perpendicular to it. Furthermore, the water-cooling procedure mostly affects the efficiency at a 45° angle. Panel efficiency increases by 0.9% at a 45° tilt angle (with cooling) and, for other tilt angles, the efficiency increment varies from 0.6% to 0.7%. However, the cooling does not have much of an effect for the lowest (0°) and highest (75°) tilt angles, as, at a 0° angle, the water cannot flow over the panel and, at a 75° angle, the water passes over the panel too quickly; hence, the water does not have enough time to extract heat from the panel. Fig. 6: Open in new tabDownload slide PV-panel efficiency with and without forced cooling One of the objectives of this study is to check the efficiency of the panel at different mass flow rates of water. With the increase in the mass flow rate, the efficiency increases proportionally as more water extracts more heat from the panel (Fig. 7). From the experiment, an efficiency of 14.9% is obtained for a mass flow rate of 9.09 × 10–4 l/min/cm2, whereas, for a mass flow rate of 1.36 × 10–3 l/min/cm2, an efficiency of 16.4% (approximately) is obtained. It is mandatory to mention here that, in this system, the mass loss of water is only due to the evaporation and the leakage of water from the panel. Due to solar radiation and the heat of the panel, some of the water is evaporated while passing over the panel, as shown in Fig. 8. The evaporation rate is higher at low mass flow rates; however, with the increase in the flow rate, the evaporation rate reduces. During the experiment, all possible and visible leakages were sealed properly and, under all conditions, it is assumed that there is no leakage in the system. Fig. 7: Open in new tabDownload slide Effect of mass flow rates on PV-panel efficiency Fig. 8: Open in new tabDownload slide Mass loss of water due to evaporation and leakage Fig. 9 exhibits the collector-efficiency variations at different angles and the mass flow rates. The mass flow rate does not affect the collector efficiency as significantly as the angles do. An overall efficiency of 60–65% is obtained in this experiment for different mass flow rates at a 10° angle. With an increase in the angle (up to 30°), the efficiency of the collector increases and, beyond this angle, the efficiency tends to decrease drastically. An average of 60% collector efficiency is obtained in the current study. However, with the increase in the mass flow rate, the efficiency of the PV panel increases. Therefore, theoretically, with the increase in the mass flow rate, the overall system efficiency should be higher. Instead, the experiment verified a different result. From experiment, it is found that the system efficiency is highest at a mass flow rate of 1.14 × 10–3 l/min/cm2 (Fig. 10). This phenomenon is due to the fact that the overall system efficiency is the combination of both the PV panel and the collector efficiency. However, it is obvious that, as the mass flow rate of the water increases, the final output water temperature decreases. Therefore, it is advisable that the output water temperature should be taken into consideration while maintaining the mass flow rate. Fig. 11 exhibits the average electrical exergetic efficiency of the PV panel with and without forced cooling. The exergetic efficiency with forced water cooling is much higher than without the cooling system. At a mass flow rate of 1.36 × 10–3 l/min/cm2, the highest exergetic efficiency of 13% was obtained. It is interesting to mention here that exergy loss reduces with the minimization of entropy generation in the system because of irreversibility [35]. Usually, higher diameter and length of the tube of a collector reduce entropy generation but, in this system, as the dimension of the collector is constant, it has no effect on the entropy. However, with the increase in the mass flow rate of the water, the entropy of the collector is reduced. Adding nano fluids to the water could probably reduce the entropy even more [37], which could be a future research perspective. From exergy and exergy loss analysis, it is clear that, with the increase in the mass flow rate, the exergetic efficiency increases and the thermal exergy of the panel reduces (Fig. 12), which was one of the major objectives of this study. Fig. 9: Open in new tabDownload slide Effect of angles and mass flow rates on collector efficiency Fig. 10: Open in new tabDownload slide Overall efficiency of the system Fig. 11: Open in new tabDownload slide Average electrical exergetic efficiency of the PV panel (with and without cooling) Fig. 12: Open in new tabDownload slide Share of overall exergy and exergy loss (a) without water and (b) the mass flow rate at 1.25 × 10–3 l/min/cm2 4 Conclusion The results show that it is comparatively easy and feasible to flow water over the top surface of a PV panel. The efficiency of the PV panel increased by ~0.8–1% for different forced mass flow rates of water. By forcing the cooling water over the top surface, there is no need for any heat-transfer material to increase the heat-transfer rate between the PV panel and the collector. The major findings of the study are: (i) Photovoltaic panel efficiency was ~14–16%. The discussed methodology increased the panel efficiency significantly. Exergy analysis showed that, at a mass flow rate of 3 l/min (1.36 × 10–3 l/min/cm2), the highest electrical exergetic efficiency of 13% was obtained. (ii) Collector efficiency was ~55–65% for different cases. (iii) The highest temperature of the hot water obtained was ~66°C for a mass flow rate of 9.09 × 10–4 l/min/cm2. (iv) The highest overall energy efficiency was obtained for a mass flow rate of 1.14 × 10–3 l/min/cm2, which was around 81%. (v) The efficiency of the whole system is approximately five times higher than that of the PV system alone. Finally, the system was validated experimentally. Conflict of Interest None declared. Nomenclature M . Water-mass flow rate (kg/s) . Ƞmodule . Efficiency of PV module . Cp Specific heat of fluid (joule/gram °C) Ƞwire Efficiency of wire ΔT Temperature difference (°C) CO Convection number Is Solar-radiation intensity (W/m2) ht Heat-transfer coefficient (W/m2k) Ƞr Reference efficiency of panel f Friction factor Β Temperature coefficient (0045°C) x Quality Tsc PV temperature (°C) fr Froude number Tr Reference temperature (°C) BO Boiling number EL Daily energy consumption Nu Nusselt number PSH Peak Sun hours K Conductivity of material Ƞsand Ƞiv Efficiency of the system components L Length of plate Sf Safety factor represents the compensation of resistive losses and PV temperature losses RaL Rayleigh number APV PV array area (m2) γ Psychometric constant (pa/K) GT Daily solar radiation (KWh/m2) V2 Wind speed (m/s) m Slope of saturation vapour pressure curve (pa/K) τ Transmission coefficient of glazing Rn Net irradiance (W/m2) α Absorption coefficient of plate Cp Heat capacity of air (J/kg/K) Tc Collector average temperature (°C) ρa Air density (kg/m3) Tp Glass plate temperature (°C) ga Momentum surface aerodynamic conductance (m/s) Isc Electrical current (A) (short circuit) σv Latent heat of vaporization (J/kg) FF Fill factor Ex Exergy rate (W) Cpf Specific heat capacity of fluid (J/kg/K) G Solar-irradiation intensity (W) εth Exergy efficiency (%) (thermal) amb Ambient εel Exergy efficiency (%) (electrical) φ Stream exergy εov Exergy efficiency (%) (Overall) h Enthalpy (J/kg) Tf Temperature of fluid (K) S Specific enthalpy (J/kg/K) I Current (A) Voc Electrical voltage (V) (open circuit) V Voltage (V) M . Water-mass flow rate (kg/s) . Ƞmodule . Efficiency of PV module . Cp Specific heat of fluid (joule/gram °C) Ƞwire Efficiency of wire ΔT Temperature difference (°C) CO Convection number Is Solar-radiation intensity (W/m2) ht Heat-transfer coefficient (W/m2k) Ƞr Reference efficiency of panel f Friction factor Β Temperature coefficient (0045°C) x Quality Tsc PV temperature (°C) fr Froude number Tr Reference temperature (°C) BO Boiling number EL Daily energy consumption Nu Nusselt number PSH Peak Sun hours K Conductivity of material Ƞsand Ƞiv Efficiency of the system components L Length of plate Sf Safety factor represents the compensation of resistive losses and PV temperature losses RaL Rayleigh number APV PV array area (m2) γ Psychometric constant (pa/K) GT Daily solar radiation (KWh/m2) V2 Wind speed (m/s) m Slope of saturation vapour pressure curve (pa/K) τ Transmission coefficient of glazing Rn Net irradiance (W/m2) α Absorption coefficient of plate Cp Heat capacity of air (J/kg/K) Tc Collector average temperature (°C) ρa Air density (kg/m3) Tp Glass plate temperature (°C) ga Momentum surface aerodynamic conductance (m/s) Isc Electrical current (A) (short circuit) σv Latent heat of vaporization (J/kg) FF Fill factor Ex Exergy rate (W) Cpf Specific heat capacity of fluid (J/kg/K) G Solar-irradiation intensity (W) εth Exergy efficiency (%) (thermal) amb Ambient εel Exergy efficiency (%) (electrical) φ Stream exergy εov Exergy efficiency (%) (Overall) h Enthalpy (J/kg) Tf Temperature of fluid (K) S Specific enthalpy (J/kg/K) I Current (A) Voc Electrical voltage (V) (open circuit) V Voltage (V) Open in new tab M . Water-mass flow rate (kg/s) . Ƞmodule . Efficiency of PV module . Cp Specific heat of fluid (joule/gram °C) Ƞwire Efficiency of wire ΔT Temperature difference (°C) CO Convection number Is Solar-radiation intensity (W/m2) ht Heat-transfer coefficient (W/m2k) Ƞr Reference efficiency of panel f Friction factor Β Temperature coefficient (0045°C) x Quality Tsc PV temperature (°C) fr Froude number Tr Reference temperature (°C) BO Boiling number EL Daily energy consumption Nu Nusselt number PSH Peak Sun hours K Conductivity of material Ƞsand Ƞiv Efficiency of the system components L Length of plate Sf Safety factor represents the compensation of resistive losses and PV temperature losses RaL Rayleigh number APV PV array area (m2) γ Psychometric constant (pa/K) GT Daily solar radiation (KWh/m2) V2 Wind speed (m/s) m Slope of saturation vapour pressure curve (pa/K) τ Transmission coefficient of glazing Rn Net irradiance (W/m2) α Absorption coefficient of plate Cp Heat capacity of air (J/kg/K) Tc Collector average temperature (°C) ρa Air density (kg/m3) Tp Glass plate temperature (°C) ga Momentum surface aerodynamic conductance (m/s) Isc Electrical current (A) (short circuit) σv Latent heat of vaporization (J/kg) FF Fill factor Ex Exergy rate (W) Cpf Specific heat capacity of fluid (J/kg/K) G Solar-irradiation intensity (W) εth Exergy efficiency (%) (thermal) amb Ambient εel Exergy efficiency (%) (electrical) φ Stream exergy εov Exergy efficiency (%) (Overall) h Enthalpy (J/kg) Tf Temperature of fluid (K) S Specific enthalpy (J/kg/K) I Current (A) Voc Electrical voltage (V) (open circuit) V Voltage (V) M . Water-mass flow rate (kg/s) . Ƞmodule . Efficiency of PV module . Cp Specific heat of fluid (joule/gram °C) Ƞwire Efficiency of wire ΔT Temperature difference (°C) CO Convection number Is Solar-radiation intensity (W/m2) ht Heat-transfer coefficient (W/m2k) Ƞr Reference efficiency of panel f Friction factor Β Temperature coefficient (0045°C) x Quality Tsc PV temperature (°C) fr Froude number Tr Reference temperature (°C) BO Boiling number EL Daily energy consumption Nu Nusselt number PSH Peak Sun hours K Conductivity of material Ƞsand Ƞiv Efficiency of the system components L Length of plate Sf Safety factor represents the compensation of resistive losses and PV temperature losses RaL Rayleigh number APV PV array area (m2) γ Psychometric constant (pa/K) GT Daily solar radiation (KWh/m2) V2 Wind speed (m/s) m Slope of saturation vapour pressure curve (pa/K) τ Transmission coefficient of glazing Rn Net irradiance (W/m2) α Absorption coefficient of plate Cp Heat capacity of air (J/kg/K) Tc Collector average temperature (°C) ρa Air density (kg/m3) Tp Glass plate temperature (°C) ga Momentum surface aerodynamic conductance (m/s) Isc Electrical current (A) (short circuit) σv Latent heat of vaporization (J/kg) FF Fill factor Ex Exergy rate (W) Cpf Specific heat capacity of fluid (J/kg/K) G Solar-irradiation intensity (W) εth Exergy efficiency (%) (thermal) amb Ambient εel Exergy efficiency (%) (electrical) φ Stream exergy εov Exergy efficiency (%) (Overall) h Enthalpy (J/kg) Tf Temperature of fluid (K) S Specific enthalpy (J/kg/K) I Current (A) Voc Electrical voltage (V) (open circuit) V Voltage (V) Open in new tab References [1] Islam MT , Shahir S, Uddin TI, et al. Current energy scenario and future prospect of renewable energy in Bangladesh . Renewable & Sustainable Energy Reviews , 2014 , 39 : 1074 – 1088 . Google Scholar Crossref Search ADS WorldCat [2] Anam K . Power crisis & its solution through renewable energy in Bangladesh . Multidisciplinary Journals in Science and Technology, Journal of Selected Areas in Renewable and Sustainable Energy (JRSE) . 2011 , 13 – 18 . https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.389.5889 ( 10 September 2020 , date last accessed). [3] Bangladesh Gazette . Renewable Energy Policy of Bangladesh 2008 . 6 November 2008 . http://www.sreda.gov.bd/d3pbs_uploads/files/policy_1_rep_english.pdf. [4] Huntzinger DN , Hutchins MJ, Gierke JS, et al. Enabling sustainable thinking in undergraduate engineering education . International Journal of Engineering Education , 2007 , 23 : 218 . Google Scholar OpenURL Placeholder Text WorldCat [5] Hossain M . Achieving sustainable energy targets in Bangladesh . UN Chronicle 52 , no. 3 , 2013 , 36 – 39 . [6] Shariar KF , Ovy EG, Hossainy TA. Closed environment design of solar collector trough using lenses and reflectors . In: World Renewable Energy Congress-Sweden. Linköping, Sweden : Linköping University Electronic Press , 2011 , 3852 – 3858 . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC [7] Khan SH , Rahman TU, Hossain H. A brief study of the prospect of solar energy in generation of electricity in Bangladesh . Cyber Journals: Multidisciplinary Journals in Science Technology, Journal of Selected Areas in Renewable Sustainable Energy , June edn. 2012 . http://www.cyberjournals.com/Papers/Jun2012/02.pdf ( 10 September 2020 , date last accessed) [8] Lasnier F , Ang TG. Solar cell technologies . In: Lasnier F, Ang TG (eds). Solar Photovoltaic Handbook . Bristol : Adam Hilger , 2017 , 3 – 66 . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC [9] Oner IV , Yesilyurt MK, Yilmaz EC, et al. Photovoltaic thermal (PVT) solar panels . International Journal of New Technology and Research , 2016 , 2 : 13 – 16 . Google Scholar OpenURL Placeholder Text WorldCat [10] Balter J , Ganem C, Discoli C. On high-rise residential buildings in an oasis-city: Thermal and energy assessment of different envelope materiality above and below tree canopy . Energy and Buildings , 2016 , 113 : 61 – 73 . Google Scholar Crossref Search ADS WorldCat [11] Markowski M , Trafczynski M, Urbaniec K. Validation of the method for determination of the thermal resistance of fouling in shell and tube heat exchangers . Energy Conversion and Management , 2013 , 76 : 307 – 313 . Google Scholar Crossref Search ADS WorldCat [12] Khanjari Y , Pourfayaz F, Kasaeian AB. Numerical investigation on using of nanofluid in a water-cooled photovoltaic thermal system . Energy Conversion and Management , 2016 , 122 : 263 – 278 . Google Scholar Crossref Search ADS WorldCat [13] Verhelst J , Van Ham G, Saelens D, et al. Model selection for continuous commissioning of HVAC-systems in office buildings: a review . Renewable and Sustainable Energy Reviews , 2017 , 76 : 673 – 686 . Google Scholar Crossref Search ADS WorldCat [14] Salem M , Ali R, Elshazly K. Experimental investigation of the performance of a hybrid photovoltaic/thermal solar system using aluminium cooling plate with straight and helical channels . Solar Energy , 2017 , 157 : 147 – 156 . Google Scholar Crossref Search ADS WorldCat [15] Abdelrazik AS , Al-Sulaiman F, Saidur R, et al. Evaluation of the effects of optical filtration and nanoPCM on the performance of a hybrid photovoltaic-thermal solar collector . Energy Conversion and Management , 2019 , 195 : 139 – 156 . Google Scholar Crossref Search ADS WorldCat [16] Su D , Jia Y, Alva G, et al. Comparative analyses on dynamic performances of photovoltaic–thermal solar collectors integrated with phase change materials . Energy Conversion and Management , 2017 , 131 : 79 – 89 . Google Scholar Crossref Search ADS WorldCat [17] Souliotis M , Arnaoutakis N, Panaras G, et al. Experimental study and Life Cycle Assessment (LCA) of Hybrid Photovoltaic/Thermal (PV/T) solar systems for domestic applications . Renewable Energy , 2018 , 126 : 708 – 723 . Google Scholar Crossref Search ADS WorldCat [18] Guarracino I , Mellor A, Ekins-Daukes NJ, et al. Dynamic coupled thermal-and-electrical modelling of sheet-and-tube hybrid photovoltaic/thermal (PVT) collectors . Applied Thermal Engineering , 2016 , 101 : 778 – 795 . Google Scholar Crossref Search ADS WorldCat [19] Fan W , Kokogiannakis G, Ma Z, et al. Development of a dynamic model for a hybrid photovoltaic thermal collector–Solar air heater with fins . Renewable Energy , 2017 , 101 : 816 – 834 . Google Scholar Crossref Search ADS WorldCat [20] Othman MYH , Hussain F. Designs of various hybrid photovoltaic-thermal (PV/T) solar collectors. In: Sayigh A (ed). Photovoltaics for Sustainable Electricity and Buildings . Cham : Springer , 2017 , 95 – 112 . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC [21] Hamdoon OM , Alomar OR, Salim BM. Performance analysis of hybrid photovoltaic thermal solar system in Iraq climate condition . Thermal Science and Engineering Progress , 2020 , 17 : 100359 . Google Scholar Crossref Search ADS WorldCat [22] Sahay A , Sethi VK, Tiwari AC, et al. A review of solar photovoltaic panel cooling systems with special reference to ground coupled central panel cooling system (GC-CPCS) . Renewable and Sustainable Energy Reviews , 2018 , 125 : 936 – 946 . Google Scholar OpenURL Placeholder Text WorldCat [23] Arefin MA . Analysis of an integrated photovoltaic thermal system by top surface natural circulation of water . Frontiers in Energy Research , 2019 , 7 : 97 . Google Scholar Crossref Search ADS WorldCat [24] García MA , Balenzategui J. Estimation of photovoltaic module yearly temperature and performance based on nominal operation cell temperature calculations . Renewable Energy , 2004 , 29 : 1997 – 2010 . Google Scholar Crossref Search ADS WorldCat [25] Kotlay P , Wenk J, Bucher K. Outdoor measurements of NOCT and its influence on energy collection . In: Proceedings of the Second World Conference on PV Solar Energy Conversion . Vienna , 1998 , 2334 – 2337 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC [26] Malik A , Damit SJ. Outdoor testing of single crystal silicon solar cells . Renewable Energy , 2003 , 28 : 1433 – 1445 . Google Scholar Crossref Search ADS WorldCat [27] Daily JW. Statistical Thermodynamics: An Engineering Approach . Cambridge : Cambridge University Press , 2018 . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC [28] Krauter S , Araújo RG, Schroer S, et al. Combined photovoltaic and solar thermal systems for facade integration and building insulation . Solar Energy , 1999 , 67 : 239 – 248 . Google Scholar Crossref Search ADS WorldCat [29] Kim D-J , Kim DH, Bhattarai S, et al. Simulation and model validation of the surface cooling system for improving the power of a photovoltaic module . Journal of Solar Energy Engineering , 2011 , 133 : 041012-1-6 . Google Scholar Crossref Search ADS WorldCat [30] Roeth J , Facchini A, Bernhard N. Optimized size and tab width in partial solar cell modules including shingled designs . International Journal of Photoenergy , 2017 , 2017 : 3609109 . Google Scholar Crossref Search ADS WorldCat [31] Ma C , Iqbal M. Statistical comparison of models for estimating solar radiation on inclined surfaces . Solar Energy , 1983 , 31 : 313 – 317 . Google Scholar Crossref Search ADS WorldCat [32] Allen RG , Pereira LS, Raes D, et al. Crop evapotranspiration: guidelines for computing crop water requirements: FAO irrigation and drainage paper 56 . Fao, Rome 1998 , 300 : D05109 . Google Scholar OpenURL Placeholder Text WorldCat [33] Trout TJ , DeJonge KC. Crop water use and crop coefficients of maize in the great plains . Journal of Irrigation Drainage Engineering , 2018 , 144 : 04018009 . Google Scholar Crossref Search ADS WorldCat [34] Tiwari G , Meraj M, Khan M, et al. Improved Hottel-Whillier-Bliss equation for N-photovoltaic thermal-compound parabolic concentrator (N-PVT-CPC) collector . Solar Energy , 2018 , 166 : 203 – 212 . Google Scholar Crossref Search ADS WorldCat [35] Sardarabadi M , Hosseinzadeh M, Kazemian A, et al. Experimental investigation of the effects of using metal-oxides/water nanofluids on a photovoltaic thermal system (PVT) from energy and exergy viewpoints . Energy , 2017 , 138 : 682 – 695 . Google Scholar Crossref Search ADS WorldCat [36] Park S , Pandey A, Tyagi V, et al. Energy and exergy analysis of typical renewable energy systems . Renewable and Sustainable Energy Reviews , 2014 , 30 : 105 – 123 . Google Scholar Crossref Search ADS WorldCat [37] Said Z , Saidur R, Rahim NA, et al. Analyses of exergy efficiency and pumping power for a conventional flat plate solar collector using SWCNTs based nanofluid . Energy , 2014 , 78 : 1 – 9 . Google Scholar Crossref Search ADS WorldCat [38] Chow TT , Pei G, Fong K, et al. Energy and exergy analysis of photovoltaic–thermal collector with and without glass cover . Applied Energy , 2009 , 86 : 310 – 316 . Google Scholar Crossref Search ADS WorldCat [39] Kazemian A , Taheri A, Sardarabadi A, et al. Energy, exergy and environmental analysis of glazed and unglazed PVT system integrated with phase change material: An experimental approach . Solar Energy , 2020 , 201 : 178 – 189 . Google Scholar Crossref Search ADS WorldCat [40] Hosseinzadeh M , Sardarabadi M, Passandideh-Fard M. Energy and exergy analysis of nanofluid based photovoltaic thermal system integrated with phase change material . Energy , 2018 , 147 : 636 – 647 . Google Scholar Crossref Search ADS WorldCat © The Author(s) 2020. Published by Oxford University Press on behalf of National Institute of Clean-and-Low-Carbon Energy This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com © The Author(s) 2020. Published by Oxford University Press on behalf of National Institute of Clean-and-Low-Carbon Energy