Pub Date : 2026-07-01Epub Date: 2026-03-13DOI: 10.1109/JPHOTOV.2026.3687922
Thales Borrely;Thien Truong;Theresa E. Saenz;Sarah A. Collins;Julian Wagner;Jennifer G. Selvidge;William E. McMahon;Sneh Sinha;Mitchell Smith;Melbs LeMieux;Brett Walker;Myles A. Steiner
The accelerated increase in demand for III–V space photovoltaics on GaAs and Ge substrates, as well as growing interests in terrestrial applications, motivate the development of cost-effective, high-throughput processing routes of these materials. Here, we assess spray-coated silver (Ag) back contact metallization as a substitute for electron-beam-evaporated metals currently used in industry. We find that the spray-coated Ag films are dense and continuous. By means of quantum efficiency, dark current–voltage, and illuminated current–voltage characterizations, we show that spray-coated GaAs and Ge solar cells perform similarly to baseline devices with electroplated Au, including under high current densities. We estimate that the thresholds for specific contact resistance below which back contacts do not significantly contribute to resistive loss are $bm{2.1times 10^{-1}Omega }cdot$cm$bm{^{2}}$ for GaAs and $bm{4.7times 10^{-2}Omega }cdot$cm$bm{^{2}}$ for Ge. We experimentally confirm that our spray-coated samples meet these requirements. Peel tests show that the adhesion of plain spray-coated Ag films to the back of p-type Ge substrates used in III–V solar cells is currently insufficient, whereas adhesion to p-type GaAs substrates is outstanding and requires no further optimization.
{"title":"Spray-Coated Silver as Backside Metal for III–V Photovoltaic Devices on GaAs and Ge Substrates","authors":"Thales Borrely;Thien Truong;Theresa E. Saenz;Sarah A. Collins;Julian Wagner;Jennifer G. Selvidge;William E. McMahon;Sneh Sinha;Mitchell Smith;Melbs LeMieux;Brett Walker;Myles A. Steiner","doi":"10.1109/JPHOTOV.2026.3687922","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2026.3687922","url":null,"abstract":"The accelerated increase in demand for III–V space photovoltaics on GaAs and Ge substrates, as well as growing interests in terrestrial applications, motivate the development of cost-effective, high-throughput processing routes of these materials. Here, we assess spray-coated silver (Ag) back contact metallization as a substitute for electron-beam-evaporated metals currently used in industry. We find that the spray-coated Ag films are dense and continuous. By means of quantum efficiency, dark current–voltage, and illuminated current–voltage characterizations, we show that spray-coated GaAs and Ge solar cells perform similarly to baseline devices with electroplated Au, including under high current densities. We estimate that the thresholds for specific contact resistance below which back contacts do not significantly contribute to resistive loss are <inline-formula><tex-math>$bm{2.1times 10^{-1}Omega }cdot$</tex-math></inline-formula>cm<inline-formula><tex-math>$bm{^{2}}$</tex-math></inline-formula> for GaAs and <inline-formula><tex-math>$bm{4.7times 10^{-2}Omega }cdot$</tex-math></inline-formula>cm<inline-formula><tex-math>$bm{^{2}}$</tex-math></inline-formula> for Ge. We experimentally confirm that our spray-coated samples meet these requirements. Peel tests show that the adhesion of plain spray-coated Ag films to the back of p-type Ge substrates used in III–V solar cells is currently insufficient, whereas adhesion to p-type GaAs substrates is outstanding and requires no further optimization.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"16 4","pages":"447-452"},"PeriodicalIF":3.7,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148517458","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 : 2026-07-01Epub Date: 2026-04-22DOI: 10.1109/JPHOTOV.2026.3680213
Huajian Ding;Zhongyuan Su;Jun Wang
Aiming at the need for accurate prediction operational temperature of N-type TOPCon bifacial double-glass photovoltaic modules, this article proposes an integrated forecasting method combining physical principles with metalearning. A metalearner model based on a Stacking ensemble framework is constructed, which integrates predictions from a thermodynamic model and a dual-branch long short-term memory (LSTM) network through a dynamic weighting mechanism. Cross-validation strategies are employed to train deep neural networks, enabling multiscale feature integration and weight self-adjustment. Experimental results achieve a root mean square error (RMSE) of only 0.95°C, representing a 66% reduction compared to the physical model (RMSE = 2.81°C) and a 14% reduction compared to the LSTM model (RMSE = 1.11°C), substantially improving the model’s adaptability in nonlinear and dynamic scenarios.
{"title":"A Physics Parameter-Guided Metalearning Framework for Temperature Prediction of Double-Sided Double-Glass TopCon Photovoltaic Modules","authors":"Huajian Ding;Zhongyuan Su;Jun Wang","doi":"10.1109/JPHOTOV.2026.3680213","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2026.3680213","url":null,"abstract":"Aiming at the need for accurate prediction operational temperature of N-type TOPCon bifacial double-glass photovoltaic modules, this article proposes an integrated forecasting method combining physical principles with metalearning. A metalearner model based on a Stacking ensemble framework is constructed, which integrates predictions from a thermodynamic model and a dual-branch long short-term memory (LSTM) network through a dynamic weighting mechanism. Cross-validation strategies are employed to train deep neural networks, enabling multiscale feature integration and weight self-adjustment. Experimental results achieve a root mean square error (RMSE) of only 0.95°C, representing a 66% reduction compared to the physical model (RMSE = 2.81°C) and a 14% reduction compared to the LSTM model (RMSE = 1.11°C), substantially improving the model’s adaptability in nonlinear and dynamic scenarios.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"16 4","pages":"583-593"},"PeriodicalIF":3.7,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148443052","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 : 2026-07-01Epub Date: 2026-03-04DOI: 10.1109/JPHOTOV.2026.3665341
A. Münzer;M. Weittenhiller;M. Bivour;S. W. Glunz;F. Meyer
In this study, we investigate the influence of the wavelength on the laser annealing of transparent conductive oxides (TCO). The main goal is to confine energy deposition to the film thickness of 70 nm to 100 nm, aiming to mitigate degradation of the underlying silicon heterojunction (SHJ) solar cell surface passivation. Addressing industrialization aspects, the experiment encompasses planar and textured surfaces, three different TCOs with distinct structural and optical properties (polycrystalline Sn-doped In2O3, polycrystalline Al-doped ZnO, amorphous F-doped In2O3) and laser wavelengths ranging from 248 nm over 308 nm to 343 nm, the latter being accesible with high-throughput systems. A comparable sheet resistance ($R_{text{sh}}$) reduction is achieved for each TCO on planar substrates, despite absorptance ranging from 33% to 75% across the studied wavelength range. Specific parameters yield $R_{text{sh}}$ lower than the hotplate reference, while the implied open-circuit voltage ($mathrm{i}V_{text{OC}}$) of the SHJ solar cell remains unaffected, indicating no induced recombination-active defects in the surface passivation. Although textured surfaces limit the parameter space for laser annealing, comparably low $R_{text{sh}}$ values to the hotplate processes without $mathrm{i}V_{text{OC}}$ reduction is achieved. Our study shows that UV ns laser annealing of TCOs on temperature-sensitive SHJ solar cells is feasible on textured surfaces and with high-throughput, low-cost laser systems, demonstrating the potential for industrial solar cell production.
{"title":"Selective Annealing of Transparent Conductive Oxides on Silicon Heterojunction Solar Cells Enabled by UV Laser Processing","authors":"A. Münzer;M. Weittenhiller;M. Bivour;S. W. Glunz;F. Meyer","doi":"10.1109/JPHOTOV.2026.3665341","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2026.3665341","url":null,"abstract":"In this study, we investigate the influence of the wavelength on the laser annealing of transparent conductive oxides (TCO). The main goal is to confine energy deposition to the film thickness of 70 nm to 100 nm, aiming to mitigate degradation of the underlying silicon heterojunction (SHJ) solar cell surface passivation. Addressing industrialization aspects, the experiment encompasses planar and textured surfaces, three different TCOs with distinct structural and optical properties (polycrystalline Sn-doped In<sub>2</sub>O<sub>3</sub>, polycrystalline Al-doped ZnO, amorphous F-doped In<sub>2</sub>O<sub>3</sub>) and laser wavelengths ranging from 248 nm over 308 nm to 343 nm, the latter being accesible with high-throughput systems. A comparable sheet resistance (<inline-formula><tex-math>$R_{text{sh}}$</tex-math></inline-formula>) reduction is achieved for each TCO on planar substrates, despite absorptance ranging from 33% to 75% across the studied wavelength range. Specific parameters yield <inline-formula><tex-math>$R_{text{sh}}$</tex-math></inline-formula> lower than the hotplate reference, while the implied open-circuit voltage (<inline-formula><tex-math>$mathrm{i}V_{text{OC}}$</tex-math></inline-formula>) of the SHJ solar cell remains unaffected, indicating no induced recombination-active defects in the surface passivation. Although textured surfaces limit the parameter space for laser annealing, comparably low <inline-formula><tex-math>$R_{text{sh}}$</tex-math></inline-formula> values to the hotplate processes without <inline-formula><tex-math>$mathrm{i}V_{text{OC}}$</tex-math></inline-formula> reduction is achieved. Our study shows that UV ns laser annealing of TCOs on temperature-sensitive SHJ solar cells is feasible on textured surfaces and with high-throughput, low-cost laser systems, demonstrating the potential for industrial solar cell production.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"16 4","pages":"391-398"},"PeriodicalIF":3.7,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11421088","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148517311","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 : 2026-07-01Epub Date: 2026-03-16DOI: 10.1109/JPHOTOV.2026.3667882
Ernest I. Obetta;Jonathan Werner;Andreas Gerber;Vladimir Smirnov;Uwe Rau;Bart E. Pieters
In this study, we present a power-centric enhancement to current collection efficiency (CCE) mapping, introducing the concept of power collection efficiency (PCE). CCE, defined as the derivative of the cell current with respect to locally generated current, quantifies current transport efficiency (ranging from 0 to 1) to the device terminals. Traditionally, current efficiency maps are generated via electroluminescence imaging. Here, we propose generating PCE maps that quantify the ratio of locally generated power within the solar cell to the extracted power at the contacts. This approach captures both current and voltage losses, providing a more comprehensive assessment of power dissipation mechanisms. We demonstrate the proposed method on a silicon heterojunction solar cell under constant illumination, highlighting its potential for detailed, spatially resolved power loss diagnostics.
{"title":"Analyzing the Power Collection Efficiency of Solar Cells by Voltage-Dependent Luminescence Imaging","authors":"Ernest I. Obetta;Jonathan Werner;Andreas Gerber;Vladimir Smirnov;Uwe Rau;Bart E. Pieters","doi":"10.1109/JPHOTOV.2026.3667882","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2026.3667882","url":null,"abstract":"In this study, we present a power-centric enhancement to current collection efficiency (CCE) mapping, introducing the concept of <italic>power collection efficiency</i> (PCE). CCE, defined as the derivative of the cell current with respect to locally generated current, quantifies current transport efficiency (ranging from 0 to 1) to the device terminals. Traditionally, current efficiency maps are generated via electroluminescence imaging. Here, we propose generating PCE maps that quantify the ratio of locally generated power within the solar cell to the extracted power at the contacts. This approach captures both current and voltage losses, providing a more comprehensive assessment of power dissipation mechanisms. We demonstrate the proposed method on a silicon heterojunction solar cell under constant illumination, highlighting its potential for detailed, spatially resolved power loss diagnostics.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"16 4","pages":"453-457"},"PeriodicalIF":3.7,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11434943","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148518766","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 : 2026-07-01Epub Date: 2026-03-11DOI: 10.1109/JPHOTOV.2026.3686878
Sarah Djabali;Shamsodin Taheri;Ana-Maria Cretu;Edris Pouresmaeil
This article presents a robust and realistic evaluation of bifacial photovoltaic models. The approach mirrors standard characterization practice, enabling assessment not only of the intrinsic model accuracy but also its practical expandability under real operating conditions. The evaluation extends beyond conventional bifacial analysis and explicitly addresses rear-only performance, often neglected in the literature. The models examined are as follows: first, standard single-diode model (SDM), second, parallel-SDM, third, photocurrent-SDM, fourth, SeriesRes-SDM, and fifth, DiodeRes-SDM. Experimental measurements using Renogy RSP220DT-G1 module were conducted under different meteorological conditions, tilt angles, and surface albedo for validation. The results indicate that the DiodeRes-SDM achieves the highest bifacial accuracy [mean absolute error (MAE): 4.89%], while the SeriesRes-SDM is a comparable, simpler alternative (5.48%). Photocurrent-SDM and Parallel-SDM are most accurate for rear-only operation but perform poorly in bifacial mode. The Standard-SDM, while the simplest, introduces higher errors (7.92%). The evaluation confirms that rear operation exhibits distinct dynamics that require the development of rear-specific models. Therefore, choosing the appropriate bifacial model based on the specific application, bifacial yield prediction, system design, or installation condition optimization, is crucial for optimizing both modeling accuracy and computational efficiency.
{"title":"Performance Evaluation of Electrical Models for Bifacial PV Modules","authors":"Sarah Djabali;Shamsodin Taheri;Ana-Maria Cretu;Edris Pouresmaeil","doi":"10.1109/JPHOTOV.2026.3686878","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2026.3686878","url":null,"abstract":"This article presents a robust and realistic evaluation of bifacial photovoltaic models. The approach mirrors standard characterization practice, enabling assessment not only of the intrinsic model accuracy but also its practical expandability under real operating conditions. The evaluation extends beyond conventional bifacial analysis and explicitly addresses rear-only performance, often neglected in the literature. The models examined are as follows: first, standard single-diode model (SDM), second, parallel-SDM, third, photocurrent-SDM, fourth, SeriesRes-SDM, and fifth, DiodeRes-SDM. Experimental measurements using Renogy RSP220DT-G1 module were conducted under different meteorological conditions, tilt angles, and surface albedo for validation. The results indicate that the DiodeRes-SDM achieves the highest bifacial accuracy [mean absolute error (MAE): 4.89%], while the SeriesRes-SDM is a comparable, simpler alternative (5.48%). Photocurrent-SDM and Parallel-SDM are most accurate for rear-only operation but perform poorly in bifacial mode. The Standard-SDM, while the simplest, introduces higher errors (7.92%). The evaluation confirms that rear operation exhibits distinct dynamics that require the development of rear-specific models. Therefore, choosing the appropriate bifacial model based on the specific application, bifacial yield prediction, system design, or installation condition optimization, is crucial for optimizing both modeling accuracy and computational efficiency.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"16 4","pages":"483-493"},"PeriodicalIF":3.7,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148519012","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 : 2026-07-01Epub Date: 2026-06-22DOI: 10.1109/JPHOTOV.2026.3700436
{"title":"IEEE Journal of Photovoltaics Information for Authors","authors":"","doi":"10.1109/JPHOTOV.2026.3700436","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2026.3700436","url":null,"abstract":"","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"16 4","pages":"C3-C3"},"PeriodicalIF":3.7,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11574688","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148442970","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}
Among emerging materials, perovskite solar cells (PSCs) have attracted significant interest due to their high efficiency in laboratory settings. However, their performance under real outdoor environments remains insufficiently explored, largely because of the limited availability of perovskite module technology and the bulky, complex, and costly nature of conventional characterization equipment. This study presents a compact, cost-effective I–V tracer designed for both indoor and outdoor PSCs characterization. The device offers accuracy comparable to laboratory instruments, enabling comprehensive performance evaluation. It extracts key parameters from I–V curves, including open-circuit voltage (VOC), short-circuit current (ISC), maximum power point voltage (VMPP), maximum power point current (IMPP), fill factor (FF), and the power conversion efficiency (PCE), supporting sweeps from VOC to ISC and vice versa. It accommodates VOC up to 3 V and ISC up to 60 mA. Additionally, the system integrates environmental sensors to monitor solar irradiance (G), cell temperature (TPV), ambient temperature (TA), and humidity (HA), facilitating precise outdoor performance analysis. Experimental validation confirms its accuracy and reliability across various conditions, bridging the gap between laboratory testing and real-world PSCs assessment.
{"title":"Design and Development of a Portable I–V Tracer for Perovskite Solar Cells","authors":"Navid Tavakoli;Asfaw Negash;Rohan Khamkar;Michael Saliba","doi":"10.1109/JPHOTOV.2026.3680141","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2026.3680141","url":null,"abstract":"Among emerging materials, perovskite solar cells (PSCs) have attracted significant interest due to their high efficiency in laboratory settings. However, their performance under real outdoor environments remains insufficiently explored, largely because of the limited availability of perovskite module technology and the bulky, complex, and costly nature of conventional characterization equipment. This study presents a compact, cost-effective <italic>I</i>–<italic>V</i> tracer designed for both indoor and outdoor PSCs characterization. The device offers accuracy comparable to laboratory instruments, enabling comprehensive performance evaluation. It extracts key parameters from <italic>I</i>–<italic>V</i> curves, including open-circuit voltage (<italic>V</i><sub>OC</sub>), short-circuit current (<italic>I</i><sub>SC</sub>), maximum power point voltage (<italic>V</i><sub>MPP</sub>), maximum power point current (<italic>I</i><sub>MPP</sub>), fill factor (FF), and the power conversion efficiency (PCE), supporting sweeps from <italic>V</i><sub>OC</sub> to <italic>I</i><sub>SC</sub> and vice versa. It accommodates <italic>V</i><sub>OC</sub> up to 3 V and <italic>I</i><sub>SC</sub> up to 60 mA. Additionally, the system integrates environmental sensors to monitor solar irradiance (<italic>G</i>), cell temperature (<italic>T</i><sub>PV</sub>), ambient temperature (<italic>T<sub>A</sub></i>), and humidity (<italic>H<sub>A</sub></i>), facilitating precise outdoor performance analysis. Experimental validation confirms its accuracy and reliability across various conditions, bridging the gap between laboratory testing and real-world PSCs assessment.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"16 4","pages":"572-582"},"PeriodicalIF":3.7,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148443087","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 : 2026-07-01Epub Date: 2026-04-07DOI: 10.1109/JPHOTOV.2026.3675751
Erling W. Eriksen;Magnus M. Nygård;Niklas Erdmann;Heine N. Riise
In this article, we investigate three distinct methods of incorporating all-sky imager (ASI) images into deep learning (DL) irradiance nowcasting. The first method relies on a convolutional neural network (CNN) to extract features directly from raw RGB images. The second method uses state-of-the-art algorithms to engineer 2-D feature maps informed by domain knowledge, e.g., cloud segmentation, the cloud motion vector, solar position, and cloud base height. These feature maps are then passed to a CNN to extract compound features. The final method relies on aggregating the engineered 2-D feature maps into time-series input. Each of the three methods was then used as part of a DL model trained on a high-frequency 29-day dataset to generate multihorizon forecasts of global horizontal irradiance up to 15 min ahead. The models were then evaluated using root-mean-squared error and skill score on seven selected days of data. Aggregated engineered ASI features as model input yielded superior forecasting performance, demonstrating that the integration of ASI images into DL nowcasting models is possible without complex spatially ordered DL architectures and inputs, underscoring opportunities for alternative image processing methods as well as the potential for improved spatial DL feature processing methods.
{"title":"Deep Learning Multihorizon Irradiance Nowcasting: A Comparative Evaluation of Three Methods for Leveraging Sky Images","authors":"Erling W. Eriksen;Magnus M. Nygård;Niklas Erdmann;Heine N. Riise","doi":"10.1109/JPHOTOV.2026.3675751","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2026.3675751","url":null,"abstract":"In this article, we investigate three distinct methods of incorporating all-sky imager (ASI) images into deep learning (DL) irradiance nowcasting. The first method relies on a convolutional neural network (CNN) to extract features directly from raw RGB images. The second method uses state-of-the-art algorithms to engineer 2-D feature maps informed by domain knowledge, e.g., cloud segmentation, the cloud motion vector, solar position, and cloud base height. These feature maps are then passed to a CNN to extract compound features. The final method relies on aggregating the engineered 2-D feature maps into time-series input. Each of the three methods was then used as part of a DL model trained on a high-frequency 29-day dataset to generate multihorizon forecasts of global horizontal irradiance up to 15 min ahead. The models were then evaluated using root-mean-squared error and skill score on seven selected days of data. Aggregated engineered ASI features as model input yielded superior forecasting performance, demonstrating that the integration of ASI images into DL nowcasting models is possible without complex spatially ordered DL architectures and inputs, underscoring opportunities for alternative image processing methods as well as the potential for improved spatial DL feature processing methods.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"16 4","pages":"594-604"},"PeriodicalIF":3.7,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148443161","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}
Since its first edition in 2008, the Metallization and Interconnection Workshop for Solar Cells (MIW) has provided a forum for experts in these fields to exchange and discuss. The latest 13th edition of MIW workshop took place on 20th and 21st of October 2025 in Berlin, Germany. In this article as an opening to the special issue of IEEE Journal of Photovoltaics we summarize the highlights of presentations as well as provide a glimpse into the emerging trends based on survey completed by the participants.
{"title":"Summary of the 13th Workshop on Metallization and Interconnection of Solar Cells","authors":"Eszter Voroshazi;Guy Beaucarne;Jan Lossen;Bernd Stannowski","doi":"10.1109/JPHOTOV.2026.3688986","DOIUrl":"https://doi.org/10.1109/JPHOTOV.2026.3688986","url":null,"abstract":"Since its first edition in 2008, the Metallization and Interconnection Workshop for Solar Cells (MIW) has provided a forum for experts in these fields to exchange and discuss. The latest 13th edition of MIW workshop took place on 20th and 21st of October 2025 in Berlin, Germany. In this article as an opening to the special issue of IEEE <sc>Journal of Photovoltaics</small> we summarize the highlights of presentations as well as provide a glimpse into the emerging trends based on survey completed by the participants.","PeriodicalId":445,"journal":{"name":"IEEE Journal of Photovoltaics","volume":"16 3","pages":"308-312"},"PeriodicalIF":2.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11548692","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148146827","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}
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：