Pub Date : 2024-06-21DOI: 10.1109/JPHOTOV.2024.3414115
Valdemar M. Cavalcante;Tiago Alves Fernandes;Renato Andrade Freitas;Fabricio Bradaschia;Marcelo Cabral Cavalcanti;Leonardo Rodrigues Limongi
The main objective of this work is to propose a global nonlinear model (GNLM), valid under varying solar irradiance ( G ) and temperature ( T ) conditions, generating characteristic curves that closely replicate the actual behavior of the evaluated modules. The proposed GNLM incorporates a technique that combines surface polynomial fitting based on numerical optimization. This integration results in the creation of unique adaptable surfaces for each parameter, providing them with flexibility. Additionally, the research also aims to investigate other nonlinear global models for photovoltaic modules and conduct a comparative study of accuracy. The proposed model demonstrated significantly superior results compared with the best model evaluated in the literature for xSi modules, with a normalized mean absolute error in power (NMAEP) percentage difference of 98.21% and a normalized root mean square deviation (NRMSD) difference of 25.60%. In contrast, for mSi modules, the results showed a slight improvement over the same model, with an NMAEP percentage difference of 26.19% and an NRMSD difference of 1.52%. Similarly, for CdTe modules, there was an NMAEP percentage difference of −0.84% and an NRMSD difference of 12.82%.
{"title":"A Global Nonlinear Model for Photovoltaic Modules Based on 3-D Surface Fitting","authors":"Valdemar M. Cavalcante;Tiago Alves Fernandes;Renato Andrade Freitas;Fabricio Bradaschia;Marcelo Cabral Cavalcanti;Leonardo Rodrigues Limongi","doi":"10.1109/JPHOTOV.2024.3414115","DOIUrl":"10.1109/JPHOTOV.2024.3414115","url":null,"abstract":"The main objective of this work is to propose a global nonlinear model (GNLM), valid under varying solar irradiance (\u0000<italic>G</i>\u0000) and temperature (\u0000<italic>T</i>\u0000) conditions, generating characteristic curves that closely replicate the actual behavior of the evaluated modules. The proposed GNLM incorporates a technique that combines surface polynomial fitting based on numerical optimization. This integration results in the creation of unique adaptable surfaces for each parameter, providing them with flexibility. Additionally, the research also aims to investigate other nonlinear global models for photovoltaic modules and conduct a comparative study of accuracy. The proposed model demonstrated significantly superior results compared with the best model evaluated in the literature for xSi modules, with a normalized mean absolute error in power (NMAEP) percentage difference of 98.21% and a normalized root mean square deviation (NRMSD) difference of 25.60%. In contrast, for mSi modules, the results showed a slight improvement over the same model, with an NMAEP percentage difference of 26.19% and an NRMSD difference of 1.52%. Similarly, for CdTe modules, there was an NMAEP percentage difference of −0.84% and an NRMSD difference of 12.82%.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"14 5","pages":"815-823"},"PeriodicalIF":2.5,"publicationDate":"2024-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141548507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-21DOI: 10.1109/JPHOTOV.2024.3414192
Benjamin Figgis;Amir Abdallah;Maulid Kivambe;Ayman Samara;Brahim Aïssa;Juan Lopez Garcia;Veronica Bermudez
The growing use of photovoltaic (PV) cleaning machines (“robots”) raises the risk of abrasion to the antireflective coating (ARC) on modules’ front glass. ARC abrasion is often studied via accelerated lab tests, however field tests are needed to achieve real-world abrasion conditions. In this study nine types of PV modules and five types of ARC coupons were subjected to 18 months of dry-brush robot cleaning in the desert climate of Doha, Qatar. Three cleaning schedules were tested: daily, weekly, and never (reference samples subject to weathering alone). Modules’ power (Pmax), current (Isc), and reflectivity changes were measured and compared between the various cleaning schedules. It was found that the abrasion resistance of PV modules varied greatly. Five kinds of module showed greater losses with more frequent cleaning, while the other four did not. Lab profilometry of the coupons similarly found large variability of the depth and quantity of scratches for different ARCs, because of the difference in ARC durability between modules, and the likelihood that different cleaning robots will vary in their harshness, it is recommended to test specific robot/module pairs in the field to be confident of their ARC degradation rate.
{"title":"Abrasion of PV Antireflective Coatings by Robot Cleaning","authors":"Benjamin Figgis;Amir Abdallah;Maulid Kivambe;Ayman Samara;Brahim Aïssa;Juan Lopez Garcia;Veronica Bermudez","doi":"10.1109/JPHOTOV.2024.3414192","DOIUrl":"10.1109/JPHOTOV.2024.3414192","url":null,"abstract":"The growing use of photovoltaic (PV) cleaning machines (“robots”) raises the risk of abrasion to the antireflective coating (ARC) on modules’ front glass. ARC abrasion is often studied via accelerated lab tests, however field tests are needed to achieve real-world abrasion conditions. In this study nine types of PV modules and five types of ARC coupons were subjected to 18 months of dry-brush robot cleaning in the desert climate of Doha, Qatar. Three cleaning schedules were tested: daily, weekly, and never (reference samples subject to weathering alone). Modules’ power (Pmax), current (Isc), and reflectivity changes were measured and compared between the various cleaning schedules. It was found that the abrasion resistance of PV modules varied greatly. Five kinds of module showed greater losses with more frequent cleaning, while the other four did not. Lab profilometry of the coupons similarly found large variability of the depth and quantity of scratches for different ARCs, because of the difference in ARC durability between modules, and the likelihood that different cleaning robots will vary in their harshness, it is recommended to test specific robot/module pairs in the field to be confident of their ARC degradation rate.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"14 5","pages":"824-829"},"PeriodicalIF":2.5,"publicationDate":"2024-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141524829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-20DOI: 10.1109/JPHOTOV.2024.3412475
{"title":"IEEE Journal of Photovoltaics Information for Authors","authors":"","doi":"10.1109/JPHOTOV.2024.3412475","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2024.3412475","url":null,"abstract":"","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"14 4","pages":"C3-C3"},"PeriodicalIF":2.5,"publicationDate":"2024-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10566072","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141435523","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-20DOI: 10.1109/JPHOTOV.2024.3412813
{"title":"Call for Papers: Special Issue on Intelligent Sensor Systems for the IEEE Journal of Electron Devices","authors":"","doi":"10.1109/JPHOTOV.2024.3412813","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2024.3412813","url":null,"abstract":"","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"14 4","pages":"699-700"},"PeriodicalIF":2.5,"publicationDate":"2024-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10566079","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141435443","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-19DOI: 10.1109/JPHOTOV.2024.3414131
Ghadeer Badran;Mahmoud Dhimish
This study presents an empirical analysis of the degradation rates in eight bifacial photovoltaic (PV) systems over the initial two years of operation, comparing glass/transparent-backsheet (G/tB) and glass/glass (G/G) configurations against traditional monofacial systems. Utilizing data from various U.K. locations, we assessed systems using RdTools for degradation rate estimation—a methodology that ensures accuracy by adjusting for environmental factors, such as soiling and irradiance variations. Our findings indicate that the sampled G/tB bifacial systems exhibit higher annual degradation rates (−1.46% to −2.30%), significantly exceeding the solar industry's average (−0.8%), while the G/G configurations in our study show comparatively lower rates (−0.90% to −1.17%). Monofacial systems maintained degradation rates closer to the industry benchmark (−0.62% to −0.94%), suggesting more stable long-term performance. This article contributes novel insights into the comparative durability and efficiency of bifacial versus monofacial PV technologies based on a specific set of systems and emphasizes the critical need for advancements in bifacial system design and material quality to improve their long-term reliability.
{"title":"Short Term Performance and Degradation Trends in Bifacial Versus Monofacial PV Systems: A U.K. Case Study","authors":"Ghadeer Badran;Mahmoud Dhimish","doi":"10.1109/JPHOTOV.2024.3414131","DOIUrl":"10.1109/JPHOTOV.2024.3414131","url":null,"abstract":"This study presents an empirical analysis of the degradation rates in eight bifacial photovoltaic (PV) systems over the initial two years of operation, comparing glass/transparent-backsheet (G/tB) and glass/glass (G/G) configurations against traditional monofacial systems. Utilizing data from various U.K. locations, we assessed systems using RdTools for degradation rate estimation—a methodology that ensures accuracy by adjusting for environmental factors, such as soiling and irradiance variations. Our findings indicate that the sampled G/tB bifacial systems exhibit higher annual degradation rates (−1.46% to −2.30%), significantly exceeding the solar industry's average (−0.8%), while the G/G configurations in our study show comparatively lower rates (−0.90% to −1.17%). Monofacial systems maintained degradation rates closer to the industry benchmark (−0.62% to −0.94%), suggesting more stable long-term performance. This article contributes novel insights into the comparative durability and efficiency of bifacial versus monofacial PV technologies based on a specific set of systems and emphasizes the critical need for advancements in bifacial system design and material quality to improve their long-term reliability.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"14 5","pages":"861-864"},"PeriodicalIF":2.5,"publicationDate":"2024-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141930747","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-06DOI: 10.1109/JPHOTOV.2024.3402231
Md. Mahmudul Hasan Shihab;Redwan N. Sajjad;Mohammad Ryyan Khan
Alongside advancements in photovoltaics (PV) technology, efficient solar farm operation (e.g., strategic panel cleaning) can help lower electricity generation costs in the field. In this work, we study the effects of season-dependent soiling and cleaning on energy yield, revenue, and profit in four locations – Dhaka (Bangladesh), Oregon City (USA), Berlin (Germany), and Rumah (Saudi Arabia). These locations cover a wide range of soiling rates, season-dependent rainfall, cleaning costs, and tariff rates for a broad techno-economic analysis of soiling-affected solar farms. While a two-month cleaning cycle, for example, apparently seems too long for soiling-prone locations (Dhaka and Rumah) – this is too frequent for Oregon City and Berlin due to their high cleaning costs. Therefore, the optimal strategy for maximum profit for the latter two locations is to never clean the panels. Even at optimal cleaning cycle (maximum revenue/profit), there is a profit-loss of 13.28%, 19.97%, 1.39%, and 42.88%, compared with the respective soiling-free rated farm output at these locations. We also show the sensitivity of output and profit to the variations in soiling rate and cleaning cost; farms in Dhaka and Rumah show strong sensitivity due to the high soiling rate and low cleaning costs.
{"title":"Effects of Local Economy and Seasonal Cleaning Cycle on Yield and Profit of Soiled Solar Farms","authors":"Md. Mahmudul Hasan Shihab;Redwan N. Sajjad;Mohammad Ryyan Khan","doi":"10.1109/JPHOTOV.2024.3402231","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2024.3402231","url":null,"abstract":"Alongside advancements in photovoltaics (PV) technology, efficient solar farm operation (e.g., strategic panel cleaning) can help lower electricity generation costs in the field. In this work, we study the effects of season-dependent soiling and cleaning on energy yield, revenue, and profit in four locations – Dhaka (Bangladesh), Oregon City (USA), Berlin (Germany), and Rumah (Saudi Arabia). These locations cover a wide range of soiling rates, season-dependent rainfall, cleaning costs, and tariff rates for a broad techno-economic analysis of soiling-affected solar farms. While a two-month cleaning cycle, for example, apparently seems too long for soiling-prone locations (Dhaka and Rumah) – this is too frequent for Oregon City and Berlin due to their high cleaning costs. Therefore, the optimal strategy for maximum profit for the latter two locations is to never clean the panels. Even at optimal cleaning cycle (maximum revenue/profit), there is a profit-loss of 13.28%, 19.97%, 1.39%, and 42.88%, compared with the respective soiling-free rated farm output at these locations. We also show the sensitivity of output and profit to the variations in soiling rate and cleaning cost; farms in Dhaka and Rumah show strong sensitivity due to the high soiling rate and low cleaning costs.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"14 4","pages":"669-678"},"PeriodicalIF":2.5,"publicationDate":"2024-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141435522","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-06DOI: 10.1109/JPHOTOV.2024.3405377
James Y. Hartley;Michael A. Shimizu;Jennifer L. Braid;Ryan Flanagan;Phillip L. Reu
Stereo high-speed video of photovoltaic modules undergoing laboratory hail tests was processed using digital image correlation to determine module surface deformation during and immediately following impact. The purpose of this work was to demonstrate a methodology for characterizing module impact response differences as a function of construction and incident hail parameters. Video capture and digital image analysis were able to capture out-of-plane module deformation to a resolution of ±0.1 mm at 11 kHz on an in-plane grid of 10 × 10 mm over the area of a 1 × 2 m commercial photovoltaic module. With lighting and optical adjustments, the technique was adaptable to arbitrary module designs, including size, backsheet color, and cell interconnection. Impacts were observed to produce an initially localized dimple in the glass surface, with peak deflection proportional to the square root of incident energy. Subsequent deformation propagation and dissipation were also captured, along with behavior for instances when the module glass fractured. Natural frequencies of the module were identifiable by analyzing module oscillations postimpact. Limitations of the measurement technique were that the impacting ice ball obscured the data field immediately surrounding the point of contact, and both ice and glass fracture events occurred within 100 μs, which was not resolvable at the chosen frame rate. Increasing the frame rate and visualizing the back surface of the impact could be applied to avoid these issues. Applications for these data include validating computational models for hail impacts, identifying the natural frequencies of a module, and identifying damage initiation mechanisms.
{"title":"Measurement of Photovoltaic Module Deformation Dynamics During Hail Impact Using Digital Image Correlation","authors":"James Y. Hartley;Michael A. Shimizu;Jennifer L. Braid;Ryan Flanagan;Phillip L. Reu","doi":"10.1109/JPHOTOV.2024.3405377","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2024.3405377","url":null,"abstract":"Stereo high-speed video of photovoltaic modules undergoing laboratory hail tests was processed using digital image correlation to determine module surface deformation during and immediately following impact. The purpose of this work was to demonstrate a methodology for characterizing module impact response differences as a function of construction and incident hail parameters. Video capture and digital image analysis were able to capture out-of-plane module deformation to a resolution of ±0.1 mm at 11 kHz on an in-plane grid of 10 × 10 mm over the area of a 1 × 2 m commercial photovoltaic module. With lighting and optical adjustments, the technique was adaptable to arbitrary module designs, including size, backsheet color, and cell interconnection. Impacts were observed to produce an initially localized dimple in the glass surface, with peak deflection proportional to the square root of incident energy. Subsequent deformation propagation and dissipation were also captured, along with behavior for instances when the module glass fractured. Natural frequencies of the module were identifiable by analyzing module oscillations postimpact. Limitations of the measurement technique were that the impacting ice ball obscured the data field immediately surrounding the point of contact, and both ice and glass fracture events occurred within 100 μs, which was not resolvable at the chosen frame rate. Increasing the frame rate and visualizing the back surface of the impact could be applied to avoid these issues. Applications for these data include validating computational models for hail impacts, identifying the natural frequencies of a module, and identifying damage initiation mechanisms.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"14 4","pages":"646-651"},"PeriodicalIF":2.5,"publicationDate":"2024-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141435484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
An accurate solar irradiance forecast is critical to the reliable operation of electrical grids with increasing integration of photovoltaic systems. This study compares short-term solar irradiance forecasts based on sky images using seven different linear machine learning algorithms. In the first step, several features are extracted from sky images, reconstructed, and next used as exogenous inputs to seven machine learning algorithms, i.e., linear regression, least absolute shrinkage and selection operator (Lasso) regression, ridge regression, Bayesian ridge (BR) regression, stochastic gradient descent (SGD), generalized linear model (GLM) regression, and random sample consensus (RANSAC). A representative dataset of three years of sky images with 1-minute resolution from 2014 to 2016 serves for comparison together with the clear sky indexes as inputs to forecast ground-level solar radiances for up to 30 minutes ahead. The results of the abovementioned algorithms are compared, where for 5 and 10 minutes ahead, Lasso has the highest accuracy with a root-mean-square error (RMSE) of 0.05 and 0.062 kW/m 2 , while for 15 to 30 minutes ahead, stochastic gradient descent provides the most accurate forecast with an RMSE of 0.067, 0.071, 0.074, and 0.076 kW/m 2 for 15, 20, 25, and 30 minutes ahead horizons, respectively. For all the time horizons, Bayesian ridge is among the three most accurate models, and RANSAC has the highest error. The results show that ground-level solar irradiance can be forecasted with a relatively low average instantaneous error ranging from 0.05 to 0.1 kW/m 2 depending on the model and forecasting horizon without imposing a too high execution time overhead, namely, less than 7 s. The accuracy of the forecast can be improved if combined with cloud detection algorithms. Overall, ridge, Bayesian ridge, and stochastic gradient descent provide more accurate forecasts for short-term horizons.
{"title":"Sky Images for Short-Term Solar Irradiance Forecast: A Comparative Study of Linear Machine Learning Models","authors":"Elham Shirazi;Ivan Gordon;Angele Reinders;Francky Catthoor","doi":"10.1109/JPHOTOV.2024.3398365","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2024.3398365","url":null,"abstract":"An accurate solar irradiance forecast is critical to the reliable operation of electrical grids with increasing integration of photovoltaic systems. This study compares short-term solar irradiance forecasts based on sky images using seven different linear machine learning algorithms. In the first step, several features are extracted from sky images, reconstructed, and next used as exogenous inputs to seven machine learning algorithms, i.e., linear regression, least absolute shrinkage and selection operator (Lasso) regression, ridge regression, Bayesian ridge (BR) regression, stochastic gradient descent (SGD), generalized linear model (GLM) regression, and random sample consensus (RANSAC). A representative dataset of three years of sky images with 1-minute resolution from 2014 to 2016 serves for comparison together with the clear sky indexes as inputs to forecast ground-level solar radiances for up to 30 minutes ahead. The results of the abovementioned algorithms are compared, where for 5 and 10 minutes ahead, Lasso has the highest accuracy with a root-mean-square error (RMSE) of 0.05 and 0.062 kW/m\u0000<sup>2</sup>\u0000, while for 15 to 30 minutes ahead, stochastic gradient descent provides the most accurate forecast with an RMSE of 0.067, 0.071, 0.074, and 0.076 kW/m\u0000<sup>2</sup>\u0000 for 15, 20, 25, and 30 minutes ahead horizons, respectively. For all the time horizons, Bayesian ridge is among the three most accurate models, and RANSAC has the highest error. The results show that ground-level solar irradiance can be forecasted with a relatively low average instantaneous error ranging from 0.05 to 0.1 kW/m\u0000<sup>2</sup>\u0000 depending on the model and forecasting horizon without imposing a too high execution time overhead, namely, less than 7 s. The accuracy of the forecast can be improved if combined with cloud detection algorithms. Overall, ridge, Bayesian ridge, and stochastic gradient descent provide more accurate forecasts for short-term horizons.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"14 4","pages":"691-698"},"PeriodicalIF":2.5,"publicationDate":"2024-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141435483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：