Table of Contents Author Guidelines Submit a Manuscript
International Journal of Photoenergy
Volume 2019, Article ID 1296065, 8 pages
https://doi.org/10.1155/2019/1296065
Research Article

Reduction of Soiling on Photovoltaic Modules by a Tracker System with Downward-Facing Standby State

1Organization for Promotion of Tenure Track, University of Miyazaki, Miyazaki 889-2192, Japan
2Faculty of Engineering, University of Miyazaki, Miyazaki 889-2192, Japan

Correspondence should be addressed to Yasuyuki Ota; pj.ca.u-ikazayim.cc@ato-y

Received 17 May 2019; Revised 5 August 2019; Accepted 6 September 2019; Published 29 October 2019

Academic Editor: Giulia Grancini

Copyright © 2019 Yasuyuki Ota et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

The radiation received by solar cells within photovoltaic modules is lower than that arriving at the module surface. One of the causes of this energy loss is soiling of the module surface. Therefore, the influence of dust adhesion on photovoltaic modules must be studied. In this study, we prepared two tracker systems: a new system and a typical system. During the night, the former can switch to a downward-facing standby state, while the latter assumes an upward-facing standby state. The soiling on the polymethylmethacrylate and glass set on the tracker systems with both standby states was evaluated for 20 months in Miyazaki, Japan. By adopting the tracker system with the downward-facing standby state, a direct transmittance that was more-than-5% higher than before was consistently obtained at 500 nm in both cases with polymethylmethacrylate and glass.

1. Introduction

As the density of solar energy is low, photovoltaic (PV) systems must be installed on a large scale to generate sufficient amounts of energy. The enormous potential can be fully exploited if the world’s deserts are made available for harvesting the solar energy. For example, the Gobi Desert, which is located at high altitudes, receives large amounts of solar radiation (4.59 kWh/m2/day) [13]. While installing PV systems in deserts, we must consider the effect of the collision and adhesion of sand on PV modules. When sand remains on the module surface, it can become firmly fixed by moisture such as dew, interrupting the transmittance of light. Therefore, this type of surface adhesion on PV modules must be prevented.

The type of dust on the PV module is related to the site conditions. PV module installed near the airport is exposed to dust from airplane exhaust, and PV module surface is subject to receiving chloride damage in a coastal area [4]. In addition, various industries produced much dust, and it has adhered on the PV module [5]. For preventing the accumulation of dust, the type of dust is an important factor.

The radiation received by the cells inside PV modules is lower than that arriving at the module surface. The causes of this energy loss include soiling on the module surface and losses due to light reflection and absorption by the materials covering the cells [5, 6]. Recently, the self-cleaning coating is used to prevent dust deposition on the PV module [7, 8]. We also developed an antisoiling coating for PV module using WO3 and silica-based material [9, 10]. However, it was necessary to add cost for preparation of coating.

A concentrator photovoltaic (CPV) system uses an optical system to collect and focus sunlight onto the solar cells. As the CPV system can only work with direct sunlight and cannot focus scattered light onto the solar cells, a significant portion of the light is lost due to scattering when the collector surfaces are soiled. By contrast, conventional, nonconcentrating PV systems can work with both direct and indirect sunlight, which implies that they are much less sensitive to soiling than CPV systems [11].

Tracker systems are necessary for CPV systems, because they use optics to concentrate sunlight and the acceptance angle of the optics is narrow. In the case of 500 sun concentration systems, the acceptance angle is less than 1° [1214]. Therefore, accurate tracking of the sun is necessary for CPV systems. In addition, tracker systems are applied to conventional nonconcentrating PV systems. In the morning and evening, the acceptance surface area of sunlight is very small for fixed PV systems. By adopting a tracker system, the output of the PV system in the morning and evening can drastically increase [15].

During the night or waiting times, PV modules on trackers should preferably be horizontal to the ground because of the effect of wind resistance. Therefore, common tracker systems adopt the upward-facing standby state during the waiting time. However, sand and dust adhere and accumulate on the PV modules during this period. In this study, we developed a tracker system with a downward-facing standby state. Common tracker systems have used linear actuator or slew drive for controlling the elevation angle. Since linear actuator has a narrow moving angle, the tracker system with a downward-facing standby state requires the use of a slew drive. It can switch to the downward-facing standby state (upside-down state) during the night and sandstorms, preventing the accumulation of dust and sand. Also, the system can switch between the two states, upward- and downward-facing standby states, without additional cost, because the upward- and downward-facing standby states are controlled by operating program. Thus far, there have been no reports discussing the effect of tracker systems with a downward-facing standby state during the waiting time. In this study, the soiling on the polymethylmethacrylate (PMMA) and glass set on the tracker systems with upward- and downward-facing standby states was evaluated. We also estimated the annual energy yield of CPV and PV systems based on the degradation of transmittance of PMMA and glass and irradiance database.

2. Methods

Figure 1 displays photographs of the tracker system that can switch to a downward-facing standby state by rotating every PV module during the night and sandstorms. To allow the modules to rotate from an upward to downward orientation, the modules were divided into two on both sides across a pillar.

Figure 1: Photographs of the tracker system that can switch to a downward-facing standby state. During sandstorms and the night, this is performed by rotating all the PV modules.

Acrylic and glass are primarily used as Fresnel lens materials for the CPV modules [4, 16]. Glass is also used as the cover glass for nearly every flat-plate PV module. In this study, acrylic PMMA (SUMIPEX-E000©, Sumitomo Chemical Co., Ltd., size: , thickness: 2 mm) and photovoltaic white glass (Optiwhite™, PILKINGTON, size: , thickness: 3 mm) plates were used as substrates.

In this study, we prepared two tracker systems: a new system and a typical system. During the night, the former can switch to a downward-facing standby state (downward system), while the latter assumes an upward-facing standby state (upward system). The typical tracker system tracks the sun during the day and faces upward at night. In contrast, the downward system tracks the sun during the day but switches to a downward-facing orientation from sunset to sunrise.

The PMMA and glass substrates were set on the two tracker systems and were exposed to outdoor conditions at the University of Miyazaki (Miyazaki, Japan, 31°49N, 131°24E). The exposure period spanned from September 30, 2014, to May 24, 2016, and data were collected every two weeks. The transmittance of the samples was measured using a spectrophotometer (JASCO, V-570) with and without an integrating sphere.

3. Results and Discussion

Figure 2 shows the transmittances of the PMMA and glass substrates on the downward and upward systems after 20 months of exposure (May 24, 2016). To consider the transmitted light for nonconcentrating PV systems, which can work with both direct and indirect light, the transmittance was measured with an integrating sphere, which allows both direct and indirect (scattered) beams to be detected, as shown in Figure 2(a). To consider the transmitted light for CPV systems, which can only work with direct light, the transmittance was measured without an integrating sphere, as shown in Figure 2(b). The detectable angle of the spectrophotometer was ±0.5°, which is nearly identical to the acceptance half-angle of CPV systems with a high concentration ratio of 1000, which is our target value [17].

Figure 2: Transmittances of the PMMA and glass substrates on the downward and upward systems after 20 months of exposure (May 24, 2016) measured (a) with and (b) without an integrating sphere.

In all cases, transmittance decreases after the full exposure period (20 months). The results indicate that the soiling during the exposure causes the transmittance to decrease, and the decrease in transmittance for both direct and indirect light is considerably lower than that for direct light. It is found that a considerable amount of irradiated light was scattered by the accumulated dust on the substrates. These results indicate that CPV systems are much more sensitive to soiling than nonconcentrating PV systems. The decrease in transmittance of the glass substrate is less than that of the PMMA substrate. The existence of hydroxyl (-OH) groups on the surface is important for the prevention of dust adhesion. The presence of electrostatic charges on the surface of the substrates is one of the main factors affecting the adhesion of dust, and the charges can be suppressed by hydroxyl groups that adsorb water on the surface [18]. A very thin layer of water is formed on the surface, which prevents the localization of electrostatic charges owing to the conductivity of water. The charges scattered by the water layer can easily be discharged into the environment [19]. Moreover, the thin layer of water makes the surface hydrophilic, causing the accumulated dust to be easily washed away by rainfall. The glass surface possesses several hydroxyl groups on its surface. Therefore, the transmittance of the glass substrate maintains a higher transmittance than the PMMA substrate.

Figure 3 shows the changes in transmittance of the PMMA and glass substrates on the downward and upward systems during the exposure at a wavelength of 500 nm. Figure 3(a) shows the transmittance measured with an integrating sphere and Figure 3(b) shows that without an integrating sphere. In all cases, the initial transmittance is identical for the downward and upward systems. As the days progress, the differences in transmittance of the substrates on the downward and upward systems (transmittance of substrate on downward system minus that on upward system: ) become clear in the first six months of exposure and, subsequently, become saturated. The decrease in transmittance for direct light (Figure 3(b)) is significantly higher than those for both direct and indirect light (Figure 3(a)). As shown in Figure 3(b), the initial transmittance of the PMMA substrate measured without an integrating sphere is 91.8% (September 30, 2014) and those on the downward and upward systems after the full exposure period (May 24, 2016) are 82.2 and 77.4%, respectively, indicating that the transmittance of the PMMA substrate on the downward system is 4.8% higher than that on the upward system after the full exposure period (: 4.8%). The initial transmittance of the glass substrate measured without an integrating sphere is 90.6% (September 30, 2014), and those on the downward and upward systems after the full exposure period are 86.8 and 81.2%, respectively, indicating that the transmittance of the glass substrate on the downward system is 5.6% higher than that on the upward system after the full exposure period (: 5.6%). The averaged from March 31, 2015 (after 6 months of exposure), to May 24, 2016 (after 20 months of exposure), for the PMMA and glass substrates are 5.2 and 6.2%, respectively. By adopting the downward tracker system, we can consistently achieve a transmittance that is higher than that of the traditional system by more than 5%.

Figure 3: Changes in transmittance of the PMMA and glass substrates on the downward and upward systems during exposure at a wavelength of 500 nm measured (a) with and (b) without an integrating sphere.

Figures 4 and 5, respectively, display photographs and optical microscope images of the (a) PMMA and (b) glass substrates on the upward and downward systems after the full exposure period (May 24, 2016). It is evident that the sample surfaces of the PMMA and glass attached to the upward system are soiled in comparison to those attached to the downward system. The soiling on the surface of the PV systems can be prevented by using the downward tracker system. In all cases, millimeter-sized dust is not observed on the surface of the substrates. The tracker systems tracked the sun and were oriented almost vertically in the morning and evening. It is considered that millimeter-sized dust falls from the surface of the substrates by gravity during the vertical orientation of the tracker systems.

Figure 4: Photographs of the (a) PMMA and (b) glass substrates on the downward and upward systems after 20 months of exposure (May 24, 2016).
Figure 5: Optical microscope images of the (a) PMMA and (b) glass substrates on the downward and upward systems after 20 months of exposure (May 24, 2016).

To estimate the performances of the PVs, we estimated the output photocurrent of the PVs using the obtained transmittance data. To estimate the photocurrent of nonconcentrating Si PVs, the solar spectrum of AM 1.5G (1000 W/m2), transmittance of the PMMA and glass substrates measured with an integrating sphere, and quantum efficiency (spectral response) of the Si solar cell [20] were multiplied. This estimation assumes Si PV modules with glass and PMMA covers. Nearly every Si PV module was covered with glass. In contrast, to estimate the photocurrent of CPVs, the photocurrent from each subcell of the InGaP/InGaAs/Ge triple-junction solar cell was calculated by taking the solar spectrum of AM 1.5D (900 W/m2), transmittance of the PMMA and glass substrates measured without an integrating sphere, and spectral response of the triple-junction solar cell [21] and multiplying them. This estimation assumes CPV modules with a PMMA Fresnel lens and silicone on glass (SOG) Fresnel lens.

The photocurrent of the PVs () can be determined by Equation (1) [22, 23]: where is the wavelength, is the quantum efficiency of the solar cells, is the transmittance, and is the photon flux of the solar spectra.

Figure 6 displays the changes in the calculated normalized photocurrent of PVs using the PMMA and glass surfaces on the downward and upward systems during the exposure for nonconcentrating Si PVs and CPVs using the InGaP/InGaAs/Ge triple-junction solar cell. In this figure, the photocurrent was normalized by the initial value on September 30, 2014.

Figure 6: Changes in the calculated normalized photocurrent of the PVs using PMMA and glass surfaces on the downward and upward systems during exposure of (a) nonconcentrating Si PV and (b) CPV using InGaP/InGaAs/Ge triple-junction solar cell.

The change in photocurrent during exposure exhibits a trend identical to the transmittance shown in Figure 3. The decrease in photocurrent of the Si PV is much lower than that of the CPV because the Si PV can work with both direct and indirect light. As shown in Figure 6(b), after the full exposure period (May 24, 2016), the normalized photocurrents of the CPV with the PMMA Fresnel lens on the downward and upward systems are 0.90 and 0.85, respectively, indicating that the normalized photocurrent on the downward system is 5% higher than that on the upward system. For the same period, the normalized photocurrents of the CPV with SOG Fresnel lens on the downward and upward systems are 0.96 and 0.90, respectively, indicating that the normalized photocurrent on the downward system is 6% higher than that on the upward system. It is found that by adopting the downward tracker system, we can achieve a more-than-5% higher photocurrent after the full exposure period.

We estimated the annual energy yield of PV and CPV systems based on the degradation of photocurrent of PMMA and glass, as shown in Figure 6. The outputs of PV and CPV systems were calculated on meteorological test data for PV systems (METPV-11) [24] and configuration of PV systems (number of junction and configuration of series and strings), and the annual energy yield was integrated them. The detailed analysis procedure of the annual energy yield is described in References [25, 26]. As the estimation results, the annual energy yield of the CPV system on the downward system was 4.01% for the PMMA Fresnel lens and 4.36% for SOG Fresnel lens higher than that on the upward system. By contrast, the annual energy yield of the PV system on the downward system was 1.40% for the PMMA cover and 1.88% for glass cover higher than that on the upward system because the PV system is much less sensitive to soiling than the CPV system.

The relationship between the rainfall during the standby state at night and the effect of the downward system was assessed as follows. The accumulated rainfall during the standby state at night (from sunset to sunrise) () for each measurement interval (accumulated for two weeks before each measurement) and the differences in transmittance at 500 nm between the substrate samples on the downward and upward tracker systems () for PMMA and glass were derived from the measured data. The relationship between the two is presented in Figure 7. Here, the transmittance was measured without an integrating sphere. The open circles represent the data after the first six months of exposure (from September 30, 2014, to March 31, 2015), while the filled circles represent the data from March 31, 2015 (after six months of exposure), to May 24, 2016 (after 20 months of exposure). As mentioned regarding Figure 3, becomes distinct after the first six months of exposure and then becomes saturated. To assess the relationship between and , the data after the first six months of exposure (filled circles) are discussed. Statistically, the filled circles exhibit a weak negative correlation to and . The correlation coefficients for the PMMA and glass substrates are −0.27 and −0.42, respectively. The correlation coefficient from −0.25 to −0.40 is defined as a weak negative correlation. When there is considerable rainfall during the standby state, the substrates on the upward tracker system are washed and becomes small.

Figure 7: Relationship between the accumulated rainfall during the standby state for each measurement interval (accumulated for two weeks before each measurement) () and the difference in transmittance of the substrates on the downward and upward systems () for (a) PMMA and (b) glass.

The environment of the test site used in this study was relatively clean as it was in a rural area and experienced significant rainfall. While installing PV systems in dusty or dry areas, by adopting the downward system, the total exposure time of the systems shortens and becomes large because of the small amount of rainfall. Eventually, the effect of the downward system becomes clearer.

4. Conclusions

A tracker system that can switch to a downward-facing standby state during the night was developed and can prevent the accumulation of dust and sand. The soiling on the PMMA and glass set on the tracker systems with upward and downward-facing standby states was evaluated for 20 months in Miyazaki, Japan.

The initial transmittance of the PMMA substrate measured without an integrating sphere was 91.8% (September 30, 2014), and those on the downward and upward systems after 20 months of exposure (May 24, 2016) were 82.2 and 77.4%, respectively. The initial transmittance of the glass substrate measured without an integrating sphere was 90.6% (September 30, 2014), and those on the downward and upward systems after the full exposure period were 86.8 and 81.2%, respectively. The averaged difference in transmittance of the substrates on the downward and upward systems () from March 31, 2015 (after six months of exposure), to May 24, 2016 (after 20 months of exposure), for the PMMA and glass substrates was 5.2 and 6.2%, respectively. By adopting the downward tracker system, we can consistently achieve a more-than-5% higher transmittance. Moreover, we estimated the output photocurrent of the PVs using the obtained transmittance data. The photocurrent of the CPV with the PMMA Fresnel lens on the downward system was 5% higher than that on the upward system after the full exposure period. The photocurrent of the CPV with silicone on glass Fresnel lens on the downward system was 6% higher than that on the upward system after the full exposure period. It is found that by adopting the downward tracker system, we can achieve a more-than-5% higher photocurrent after the full exposure period.

The environment of the test site used in this study was relatively clean as it was in a rural area and not in an industrial area. The effect of the tracker system with the downward-facing standby state would be even stronger for PV systems installed at dusty sites, and the gain in the PV output would concomitantly be larger.

Data Availability

The transmittance data used to support the findings of this study are included within the article.

Disclosure

Some of the results and findings were presented in the 27th International Photovoltaic Science and Engineering Conference, Shiga, Japan, November 2017; however, detailed data and discussions were presented in this work.

Conflicts of Interest

The authors declare no conflict of interest.

Acknowledgments

This work was supported in part by a grant for the Scientific Research on Priority Areas from the University of Miyazaki. We would like to thank Mr. Shota Kurogi of the University of Miyazaki and Mr. Jun Hirota of THK Co., Ltd. for valuable discussions.

References

  1. A. Adiyabat, K. Kurokawa, K. Otani et al., “Evaluation of solar energy potential and PV module performance in the Gobi Desert of Mongolia,” Progress in Photovoltaics: Research and Applications, vol. 14, no. 6, pp. 553–566, 2006. View at Publisher · View at Google Scholar · View at Scopus
  2. M. Ito, K. Kato, H. Sugihara, T. Kichimi, J. Song, and K. Kurokawa, “A preliminary study on potential for very large-scale photovoltaic power generation (VLS-PV) system in the Gobi desert from economic and environmental viewpoints,” Solar Energy Materials and Solar Cells, vol. 75, no. 3-4, pp. 507–517, 2003. View at Publisher · View at Google Scholar · View at Scopus
  3. M. Ito, K. Kato, K. Komoto, T. Kichimi, and K. Kurokawa, “A comparative study on cost and life‐cycle analysis for 100 MW very large‐scale PV (VLS‐PV) systems in deserts using m‐Si, a‐Si, CdTe, and CIS modules,” Progress in Photovoltaics: Research and Applications, vol. 16, no. 1, pp. 17–30, 2008. View at Publisher · View at Google Scholar · View at Scopus
  4. K. Araki, T. Yano, and Y. Kuroda, “30 kW concentrator photovoltaic system using dome-shaped Fresnel lenses,” Optics Express, vol. 18, no. S1, pp. A53–A63, 2010. View at Publisher · View at Google Scholar · View at Scopus
  5. H. K. Elminir, A. E. Ghitas, R. H. Hamid, F. el-Hussainy, M. M. Beheary, and K. M. Abdel-Moneim, “Effect of dust on the transparent cover of solar collectors,” Energy Conversion and Management, vol. 47, no. 18-19, pp. 3192–3203, 2006. View at Publisher · View at Google Scholar · View at Scopus
  6. M. García, L. Marroyo, E. Lorenzo, and M. Pérez, “Soiling and other optical losses in solar‐tracking PV plants in navarra,” Progress in Photovoltaics: Research and Applications, vol. 19, no. 2, pp. 211–217, 2011. View at Publisher · View at Google Scholar · View at Scopus
  7. R. Appels, B. Lefevre, B. Herteleer et al., “Effect of soiling on photovoltaic modules,” Solar Energy, vol. 96, pp. 283–291, 2013. View at Publisher · View at Google Scholar · View at Scopus
  8. L.-Z. Zhang, A.-J. Pan, R.-R. Cai, and H. Lu, “Indoor experiments of dust deposition reduction on solar cell covering glass by transparent super-hydrophobic coating with different tilt angles,” Solar Energy, vol. 188, pp. 1146–1155, 2019. View at Publisher · View at Google Scholar
  9. K. Nabemoto, Y. Sakurada, Y. Ota et al., “Effect of anti-soiling layer coated on poly(methyl methacrylate) for concentrator photovoltaic modules,” Japanese Journal of Applied Physics, vol. 51, no. 10S, article 10ND11, 2012. View at Publisher · View at Google Scholar
  10. T. Hirohata, Y. Ota, and K. Nishioka, “Anti-soiling coating based on silica for Fresnel lens of concentrator photovoltaics,” Japanese Journal of Applied Physics, vol. 54, no. 8S1, 2015. View at Publisher · View at Google Scholar · View at Scopus
  11. M. Vivar, R. Herrero, I. Antón et al., “Effect of soiling in CPV systems,” Solar Energy, vol. 84, no. 7, pp. 1327–1335, 2010. View at Publisher · View at Google Scholar · View at Scopus
  12. K. Araki, H. Uozumi, T. Egami et al., “Development of concentrator modules with dome‐shaped Fresnel lenses and triple‐junction concentrator cells,” Progress in Photovoltaics: Research and Applications, vol. 13, no. 6, pp. 513–527, 2005. View at Publisher · View at Google Scholar · View at Scopus
  13. P. Benítez, J. C. Miñano, P. Zamora et al., “High performance Fresnel-based photovoltaic concentrator,” Optics Express, vol. 18, no. S1, pp. A25–A40, 2010. View at Publisher · View at Google Scholar · View at Scopus
  14. M. Victoria, S. Askins, R. Herrero, I. Antón, and G. Sala, “Assessment of the optical efficiency of a primary lens to be used in a CPV system,” Solar Energy, vol. 134, pp. 406–415, 2016. View at Publisher · View at Google Scholar · View at Scopus
  15. N. Shibata, Y. Ota, Y. Sakurada et al., “Output comparison of CPV and flat-plate systems in Japanese meteorological condition,” in AIP Conference Proceedings, vol. 1407, pp. 341–344, Las Vegas, NV, USA, December 2011. View at Publisher · View at Google Scholar · View at Scopus
  16. M. Steiner, G. Siefer, T. Schmidt, M. Wiesenfarth, F. Dimroth, and A. W. Bett, “43% sunlight to electricity conversion efficiency using CPV,” IEEE Journal of Photovoltaics, vol. 6, no. 4, pp. 1020–1024, 2016. View at Publisher · View at Google Scholar · View at Scopus
  17. Y. Ota and K. Nishioka, “Tracking error analysis of concentrator photovoltaic module using total 3-dimensional simulator,” in AIP Conference Proceedings, vol. 1407, pp. 281–284, Las Vegas, NV, USA, December 2011. View at Publisher · View at Google Scholar · View at Scopus
  18. Y. Horiuchi and H. Yamashita, “Design of mesoporous silica thin films containing single-site photocatalysts and their applications to superhydrophilic materials,” Applied Catalysis A: General, vol. 400, no. 1-2, pp. 1–8, 2011. View at Publisher · View at Google Scholar · View at Scopus
  19. T. Sueto, Y. Ota, and K. Nishioka, “Suppression of dust adhesion on a concentrator photovoltaic module using an anti-soiling photocatalytic coating,” Solar Energy, vol. 97, pp. 414–417, 2013. View at Publisher · View at Google Scholar · View at Scopus
  20. M. A. Green, K. Emery, Y. Hishikawa, W. Warta, and E. D. Dunlop, “Solar cell efficiency tables (version 39),” Progress in Photovoltaics: Research and Applications, vol. 20, no. 1, pp. 12–20, 2012. View at Publisher · View at Google Scholar · View at Scopus
  21. M. A. Green, K. Emery, Y. Hishikawa et al., “Solar cell efficiency tables (version 49),” Progress in Photovoltaics: Research and Applications, vol. 25, no. 1, pp. 3–13, 2017. View at Publisher · View at Google Scholar · View at Scopus
  22. D. Bouhafs, A. Moussi, A. Chikouche, and J. M. Ruiz, “Design and simulation of antireflection coating systems for optoelectronic devices: application to silicon solar cells,” Solar Energy Materials and Solar Cells, vol. 52, no. 1-2, pp. 79–93, 1998. View at Publisher · View at Google Scholar · View at Scopus
  23. Y. Ota and K. Nishioka, “Three-dimensional simulating of concentrator photovoltaic modules using ray trace and equivalent circuit simulators,” Solar Energy, vol. 86, no. 1, pp. 476–481, 2012. View at Publisher · View at Google Scholar · View at Scopus
  24. A. Itagaki, H. Okamura, and M. Yamada, “Preparation of meteorological data set throughout japan for suitable design of pv systems,” in 3rd World Conference on Photovoltaic Energy Conversion, pp. 2074–2077, Osaka, Japan, May 2003.
  25. K. Araki, Y. Ota, K.-H. Lee, T. Sakai, K. Nishioka, and M. Yamaguchi, “Analysis of fluctuation of atmospheric parameters and its impact on performance of CPV,” in AIP Conference Proceedings, vol. 2012, Puertollano, Spain, September 2018. View at Publisher · View at Google Scholar · View at Scopus
  26. H. Saiki, T. Sakai, K. Araki et al., “Verification of uncertainty in CPV’s outdoor performance,” in 7rd World Conference on Photovoltaic Energy Conversion, Waikoloa Village, HI, USA, June 2018. View at Publisher · View at Google Scholar · View at Scopus