Experimental investigation of passively cooled photovoltaic modules on the power output performance

Emy Zairah Ahmad, Kamaruzzaman Sopian, Adnan Ibrahim, Chin Kim Gan, Mohd Suffian Ab Razak Solar Energy Research Institute, Universiti Kebangsaan Malaysia, Bangi Selangor, Malaysia Faculty of Electrical & Electronic Engineering Technology, Universiti Teknikal Malaysia Melaka, Malaysia CeRIA Faculty of Electrical Engineering, Universiti Teknikal Malaysia Melaka, Malaysia Faculty of Mechanical and Manufacturing Engineering Technology, Universiti Teknikal Malaysia Melaka, Malaysia


INTRODUCTION
Solar energy has the potential to reduce the profound long-term threat of global carbon dioxide (CO2) emissions. Research on alternative energy resources started to emerge, and evaluating each technology is necessary to support climate change policies and mitigation. The International Energy Agency (IEA) has recently reported that 27% of global energy generation in 2050 will be supplied by solar photovoltaic (PV). However, an in-depth research is required to enable the competitiveness of solar PV in the future energy market. Nevertheless, a silicon (Si) PV module technology absorbs 80 % of incident solar irradiance, but only 20 % is converted into useful energy [1]. The percentage of power reduction per degree celsius ranges typically from 0.2%/⁰C to 0.5%/⁰C and highly influences the decrease in voltage and power conversion efficiency (PCE) [2]. It has been a constant challenge for a PV system exposed to continuously high solar irradiance [3]. To overcome this limitation, keeping the module temperature low is necessary.
Various authors reported significant studies on different PV cooling methodologies, and it can be categories as passive and active cooling [4]. The primary concern of PV cooling is to eliminate excess heat in periods of high solar irradiance and over long field exposures. Despite various cooling techniques addressed in [5]- [11], passive cooling is more promising than active cooling. They require no auxiliary input power, and surrounding air naturally cools the PV modules. For instance, Johnston et al. [12] proposed a continuous  [13], [14] with rectangular fin profiles tested using 250 Wp poly-Si modules. The authors concluded that, at 100 mm fin height, the effect of fin thickness is not noticeable. The proposed heat sinks have reduced the module temperature by 4 ⁰C, with an improvement of 3 % in the overall power conversion efficiency. In addition, Elbreki et al. [15] proposed a novel lapping fin with planar reflectors and tested it using a 40 Wp poly-Si module. The experimental results demonstrated that the lapping fin profiles outweigh the rectangular fin of similar fin height (200 mm) with a reduced temperature of 24.6 ⁰C. The authors also reported that the module electrical efficiency with lapping fins with planar reflector increased 10.68%. Cabo et al. [16] conducted studies with randomly positioned perforated fins to reduce the module temperature. The proposed fins were tested on a 50 Wp poly-Si module under outdoor testing conditions, and the fin height was maintained at 100 mm. The experimental results showed a 2% relative increase in electrical efficiency. Arifin et al. [17] experimentally analyse the effect of using perforated fins on PV module temperature. The average module temperature with and without perforated heat sinks was recorded at 72.8 ⁰C and 85.3 ⁰C, respectively. The overall maximum power was increased by 18.67%. Bayrak et al. [18] studied module temperature using staggered vertical fins. The highest temperature reduction of 3.39 ⁰C was observed under the 772.83 W/m 2 solar irradiance, accounting for an 11.55% efficiency improvement. On the other hand, Perez et al. [19] proposed alternative fin geometries with an angled-discontinuous fin profile and improved the heat extraction. The experimental observations indicated that the temperature reduction was within the range of 5-7 ⁰C and increased power yield to 2.96%. Selimefendigil et al. [20] studied porous aluminum foams were applied as cooling fins. The foam thickness was variated from 6 to 10 mm. The results showed that porous aluminum foams had reduced the PV temperature up to 1 ⁰C and the effect is not significant. Since the passive cooling technique showed promising findings, there is a strong need for further improvement under realistic conditions to bridge the gap between the research and industrial needs. The recent developments on PV cooling by the various authors with heat sinks are illustrated in Figure 1. [12] [  Since the output performances of PV modules are site-dependent, it is essential to investigate the influence of cooling heat sinks under outdoor testing conditions. Therefore, the proposed passive cooling technique under Malaysian climates was conducted by outdoor experimental means. The primary motivation of this study is to investigate the impact of cooling heat sinks on the electrical output performance of commercialized PV modules at a specific geographical condition. Thus, the findings of this study are proved reliable for improvement in temperature reduction using fin heat sinks. It allows stakeholders such as PV installers, PV plant owners, and other parties to favor an informed decision towards the heat sink cooling mechanism to enhance the PV system performance.

METHODOLOGY
The electrical output performance of a PV module is highly site-dependent, but the rated nominal power is based on the controlled environments known as standard test conditions. Various methods have been proposed to determine the module efficiency through simplified working equations [21]- [24]. However, the following equation is widely used to determine the PV module efficiency described as follows (1) [25].
is the efficiency of a PV module at = (25⁰C), G is the irradiance level measured in W/m 2 , and TPV is the PV module temperature. and are the coefficients for the temperature and solar irradiance, respectively. The values for , , and are given in the module datasheet. However, under real operating conditions (ROC), the power output of the PV modules installed at the test site differs from the power stated in the module datasheet. Several derating factors that need to be considered, known as power derating, kpower derating, and can be determined using (2) and (3) Where Pstc is the power rated at STC as per manufacturer's specification (Wp), kmodule_mismatch is the module mismatch derating factor, ktemp is the module temperature derating factor, kg is the peak sun factor obtained by dividing the instantaneous irradiance with 1000 W/m 2 , kdust is the dust effect derating factor, and kaging is the derating factor due to PV module's aging. In this study, kmodule_mismatch, kdust, and kaging parameters are constant (=1.0) throughout the experiment and can be combined as α. Therefore, the (2) becomes; The values of ktemp and kg can be estimated using [27]: where is the module's temperature coefficient given in %/⁰C. The module temperature, Tmodule is measured using K-type thermocouples positioned at the front and backside of PV modules. The IV curve defines the performance of PV modules, which shows how the current varies as a function of the voltage. Based on the IV curve, several parameters are used to characterize the electrical performance of the PV module, as presented in the fill factor (FF) in (7).
Where Imp is current recorded at the maximum point of the IV curve, Isc is short-circuit current, Vmp is the voltage at maximum power, and Voc is open-circuit voltage [27]. The FF can be used to determine the efficiency of a PV module at specific irradiance and temperature. The expected yield can be defined as (8) [28]. 523 two identical monocrystalline PV modules (120 Wp) and tilted at 18⁰ from the horizontal facing South to receive the maximum solar radiation. The fin geometry is illustrated in Figure 2 (a). One module was attached with the heat sinks for cooling (Panel A), while the other was used as a reference module (Panel B), as demonstrated in Figure 2 (b). The measures of incident solar irradiance shall comply with IEC 60904-3 [29]. Therefore, this work uses PV reference cells of the same glazing and technology for solar irradiance measurement with ±0.2% sensitivity. All measurements were recorded every 2s interval from 09:00 a.m to 05:00 p.m. The K-type thermocouples were used to measure the ambient temperature and the surface temperature of PV modules. The thermocouples were placed at several points at the front and rear side of PV modules having a sensitivity of ±0.5% and the temperature range of -200 ⁰C to +300 ⁰C. All temperature sensors were connected to the advanced data acquisition modules (ADAM 4018+), interfaced with the single RS-485 network.
(a) (b) Figure 2. The detailed illustrations of (a) proposed fins geometry and (b) the experimental setup

The influence of fins heat sink on the PV module temperature
The two modules were installed to investigate comparatively the PV-Fin module with the conventional module. Although all the data was recorded for several days, the data is presented for the day with clear sky and stable solar irradiance on July 19, 2021. Solar irradiance, ambient temperature, and the surface temperature of PV modules were measured at the test site and presented in Figure 3 and Figure 4. Since Malaysia is located near the equator, the climate is hot and humid throughout the year. The recorded temperatures were high and stable between 26 to 37 ⁰C throughout the day. It should be noted from the solar irradiance plot that cloudy conditions have occurred in the morning from 10:00 a.m. to 3:00 p.m., characterized by the dotted lines. The average highest and lowest solar irradiance values were recorded at 980 and 520 W/m 2 , respectively.
As shown in Figure 4, the graph for the surface temperature of PV modules (with fins and without fins) is highly influenced by solar irradiance. The PV surface temperature plot is observed to have similar trends with the plot of solar irradiance. The surface temperature of the PV module with fins is lower than the surface temperature of the PV module without fins throughout the day. The maximum temperature difference between the two modules is 5.77 ⁰C, and the average temperature difference is 3.25 ⁰C. The surface temperature plot shows one peculiarity: The module with fin demonstrates a significant impact on the module temperature reduction at solar irradiance greater than 600 W/m 2 . The surface temperature reduction was observed between 3.47 and 4.60 ⁰C. Meanwhile, the temperature reduction between both modules was insignificance as the weather got too cloudy, as shown in areas X and Y. The temperature reduction drops between 1.29 to 1.5 ⁰C. The temperature derating factor is an important parameter for assessing the PV module performance under long-term field exposure. Hence, the amount of ktemp for both modules was determined based on (5). The lowest recorded ktemp for PV with fins and PV without fins were 0.908 and 0.917, respectively. Based on Figure 5, it can be concluded that fin heat sinks positively influenced the ktemp by at least 3.4 % throughout the day.   525 current at maximum power (Imp), and voltage at maximum power (Vmp). During these measurements, the minimum and maximum recorded solar irradiances were 520 and 980 W/m 2 , respectively. It can be observed that the Voc for the PV module with fins is higher than the reference module by at least 3.14 % (see Figure 6). However, Figure 7 shows the power curves for both PV modules tested under outdoor operating conditions. Since solar cells are made up of semiconductor materials, the increase in temperature excites excess electrons and holes, causing the greater depletion region width known as the charge separation layer [30]. Hence, the Voc is very much dependent on the temperature. Besides, the value of Vmp had increased from 17.2 V to 20.4 V when using fins. Consequently, the fill factor (FF) values had improved from 0.744 to 0.826. The recorded Isc for the PV module with fins and without fins was 6.08 A and 6.10 A, respectively. The change of Isc with temperature highly depends on the light trapping properties of the designed solar cell. Hence, it can be observed that the change in Isc is much smaller than the VOC. The overall electrical performances for tested PV modules are summarized in Table 1. It is worth noting that when the module temperature drops, the voltage rises, resulting in a substantial increase in available maximum electrical power despite a slight decrease in short-circuit current. The average power output is improved by 14.2 % when using fin heat sinks as the cooling approach, as shown in Figure 8.

CONCLUSION
This work reports the experimental findings on the passively cooled PV modules using rectangular fin heat sinks under outdoor testing conditions. It can be concluded that the proposed cooling approach results in a 3.25 ⁰C average reduction in the PV module temperature. Thus, the maximum electrical power output increases up to 14.2 % by integrating the fin heat sink at the backside of the PV module. It was found that the heat sink effect on the PV module performance is significant at high solar irradiance than low solar irradiance. The finding suggests the feasibility of implementing the heat sink as a cooling approach in countries near the equator.