Todd Karin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Todd Karin has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Anubhav Jain
Energy Technologies Area, Lawrence Berkeley National Laboratory, Berkeley, USA
5 shared publications - David C. Miller
National Renewable Energy Laboratory (NREL), Golden, CO (United States)
2 shared publications - Michael G. Deceglie
National Renewable Energy Laboratory
2 shared publications - Peter Hacke
National Renewable Energy Laboratory, Golden, CO 80401, USA
2 shared publications - Adam B. Shinn
Wh Analytics, San Francisco, CA, United States
1 shared publication - Archana Sinha
SLAC National Accelerator Laboratory, Menlo Park, CA, USA
1 shared publication
Latest Publications
- 20 March 2023
- Published by Elsevier in Solar Energy
- Vol. 254, 168-181
Determining circuit model parameters from operation data for PV system degradation analysis: PVPRO
- 20 March 2023
- Published by Elsevier in Solar Energy
- Vol. 254, 168-181
Determining circuit model parameters from operation data for PV system degradation analysis: PVPRO
- 10 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Photovoltaics
- Vol. 13 (3) , 334-342
Automatic Crack Segmentation and Feature Extraction in Electroluminescence Images of Solar Modules
The effect of cracks in solar cells on the long-term degradation of photovoltaic (PV) modules remains to be determined. To investigate this effect in future studies, it is necessary to quantitatively describe the crack features (e.g., length) and correlate them with module power loss. Electroluminescence (EL) imaging is a common technique for identifying cracks. However, it is currently challenging and time-consuming to identify cracks in a large number of EL images and quantify complex crack features by human inspection. This article introduces a fast semantic segmentation method ( $\sim$ 0.18 s/cell) to automatically segment cracks from EL images and algorithms to extract crack features. We fine-tuned a UNet neural network model using pretrained VGG16 as the encoder and obtained an average F1 score of 0.875 and an intersection over union score of 0.782 on the testing set. With cracks and busbars segmented, we developed algorithms for extracting crack features, including the crack-isolated area, the brightness inside the isolated area, and the crack length. We also developed an automatic preprocessing tool for cropping individual cell images from EL images of PV modules ( $\sim$ 0.72 s/module). Our codes are published as open-source an software, and our annotated dataset composed of various types of cells is published as a benchmark for crack segmentation in EL images.
- 10 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Photovoltaics
- Vol. 13 (3) , 334-342
Automatic Crack Segmentation and Feature Extraction in Electroluminescence Images of Solar Modules
The effect of cracks in solar cells on the long-term degradation of photovoltaic (PV) modules remains to be determined. To investigate this effect in future studies, it is necessary to quantitatively describe the crack features (e.g., length) and correlate them with module power loss. Electroluminescence (EL) imaging is a common technique for identifying cracks. However, it is currently challenging and time-consuming to identify cracks in a large number of EL images and quantify complex crack features by human inspection. This article introduces a fast semantic segmentation method ( $\sim$ 0.18 s/cell) to automatically segment cracks from EL images and algorithms to extract crack features. We fine-tuned a UNet neural network model using pretrained VGG16 as the encoder and obtained an average F1 score of 0.875 and an intersection over union score of 0.782 on the testing set. With cracks and busbars segmented, we developed algorithms for extracting crack features, including the crack-isolated area, the brightness inside the isolated area, and the crack length. We also developed an automatic preprocessing tool for cropping individual cell images from EL images of PV modules ( $\sim$ 0.72 s/module). Our codes are published as open-source an software, and our annotated dataset composed of various types of cells is published as a benchmark for crack segmentation in EL images.
- 6 June 2022
- Published by Wiley in Progress in Photovoltaics: Research and Applications
- Vol. 30 (11) , 1270-1288
Photovoltaic module antireflection coating degradation survey using color microscopy and spectral reflectance
Commercial antireflective coatings (ARCs) on photovoltaic (PV) module glass can improve module power by 2.5%–3.0%, but their long‐term field performance requires additional study. In this paper, we investigate ARC performance on fielded modules using two nondestructive techniques: reflectance spectroscopy and RGB microscopy, finding a large variation in coating durability and performance. For new coatings, the fleet average nominal power enhancement is 2.8%. This power enhancement is observed to degrade at approximately −0.05%/year absolute over the first 8 years of fielding in arrays that are not regularly cleaned. Interferometry and imaging results imply that coating loss for these modules is due to slow chemical thinning of the initial 125 nm thick coating by 1.4 to 5.0 nm/year. Due to the physics of interference coatings, the performance loss is projected to accelerate as the coating is further thinned, resulting in a coating lifetime (time to 80% of initial performance) of 7.5 to 25 years. At the higher 5.0 nm/year coating loss rates, we estimate that −0.14%/year power degradation of the module can be attributed solely to ARC degradation over the first 20 years. Extreme coating loss is observed on some modules where after 8 years fielding, the coating can be completely removed with a single wet wipe with a lens tissue. A case study is also presented comparing ARC and noncoated modules installed at a single site in 2013. We find that the ARC modules have a 2.6% higher soiling power loss than the noncoated modules; this exceeds the ARC power enhancement of 2.4% and leads to the surprising conclusion that the ARC lowers production for modules in this location due to increased soiling. These results demonstrate that RGB microscopy is a powerful, field‐capable, and quantitative characterization technique for assessing degradation of PV module ARCs.
- 6 June 2022
- Published by Wiley in Progress in Photovoltaics: Research and Applications
- Vol. 30 (11) , 1270-1288
Photovoltaic module antireflection coating degradation survey using color microscopy and spectral reflectance
Commercial antireflective coatings (ARCs) on photovoltaic (PV) module glass can improve module power by 2.5%–3.0%, but their long‐term field performance requires additional study. In this paper, we investigate ARC performance on fielded modules using two nondestructive techniques: reflectance spectroscopy and RGB microscopy, finding a large variation in coating durability and performance. For new coatings, the fleet average nominal power enhancement is 2.8%. This power enhancement is observed to degrade at approximately −0.05%/year absolute over the first 8 years of fielding in arrays that are not regularly cleaned. Interferometry and imaging results imply that coating loss for these modules is due to slow chemical thinning of the initial 125 nm thick coating by 1.4 to 5.0 nm/year. Due to the physics of interference coatings, the performance loss is projected to accelerate as the coating is further thinned, resulting in a coating lifetime (time to 80% of initial performance) of 7.5 to 25 years. At the higher 5.0 nm/year coating loss rates, we estimate that −0.14%/year power degradation of the module can be attributed solely to ARC degradation over the first 20 years. Extreme coating loss is observed on some modules where after 8 years fielding, the coating can be completely removed with a single wet wipe with a lens tissue. A case study is also presented comparing ARC and noncoated modules installed at a single site in 2013. We find that the ARC modules have a 2.6% higher soiling power loss than the noncoated modules; this exceeds the ARC power enhancement of 2.4% and leads to the surprising conclusion that the ARC lowers production for modules in this location due to increased soiling. These results demonstrate that RGB microscopy is a powerful, field‐capable, and quantitative characterization technique for assessing degradation of PV module ARCs.
- Open Access
- 20 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1606-1609
Assessing Machine Learning as a Tool to Explain Variance in Deployed Photovoltaic (PV) System Degradation
Degradation remains a large uncertainty in forecasting production for PV plants, creating significant risk for developers and financiers. This study aims to quantify the distribution and drivers of degradation across 10,000 PV systems deployed for distributed or utility generation by training a machine learning model to predict year-over-year degradation rates from metadata characteristics. A combination of K-Means clustering and random forest regressor were found to associate multiple metadata features as potential drivers of degradation, including module characteristics, system design, and climate features. From this, it is inferred that if machine learning is able to find complex patterns between metadata features and system performance loss, such methods can be employed to help developers and financiers make data-informed decisions when estimating long-term energy production forecasts in financial models.
- 20 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1606-1609
Open AccessAssessing Machine Learning as a Tool to Explain Variance in Deployed Photovoltaic (PV) System Degradation
Degradation remains a large uncertainty in forecasting production for PV plants, creating significant risk for developers and financiers. This study aims to quantify the distribution and drivers of degradation across 10,000 PV systems deployed for distributed or utility generation by training a machine learning model to predict year-over-year degradation rates from metadata characteristics. A combination of K-Means clustering and random forest regressor were found to associate multiple metadata features as potential drivers of degradation, including module characteristics, system design, and climate features. From this, it is inferred that if machine learning is able to find complex patterns between metadata features and system performance loss, such methods can be employed to help developers and financiers make data-informed decisions when estimating long-term energy production forecasts in financial models.
- 23 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Photovoltaics
- Vol. 11 (3) , 760-769
Nondestructive Characterization of Antireflective Coatings on PV Modules
Antireflective coatings (ARCs) are used on the vast majority of solar photovoltaic (PV) modules to increase power production. However, ARC longevity can vary from less than 1 to over 15 years depending on coating quality and deployment conditions. A technique that can quantify ARC degradation nondestructively on commercial modules would be useful both for in-field diagnostics and accelerated aging tests. In this article, we demonstrate that accurate measurements of ARC spectral reflectance can be performed using a modified commercially available integrating-sphere probe. The measurement is fast, accurate, nondestructive, and can be performed outdoors in full-sun conditions. We develop an interferometric model that estimates coating porosity, thickness, and fractional area coverage from the measured reflectance spectrum for a uniform single-layer coating. We demonstrate the measurement outdoors on an active PV installation, identify the presence of an ARC, and estimate the properties of the coating.
- 23 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Photovoltaics
- Vol. 11 (3) , 760-769
Nondestructive Characterization of Antireflective Coatings on PV Modules
Antireflective coatings (ARCs) are used on the vast majority of solar photovoltaic (PV) modules to increase power production. However, ARC longevity can vary from less than 1 to over 15 years depending on coating quality and deployment conditions. A technique that can quantify ARC degradation nondestructively on commercial modules would be useful both for in-field diagnostics and accelerated aging tests. In this article, we demonstrate that accurate measurements of ARC spectral reflectance can be performed using a modified commercially available integrating-sphere probe. The measurement is fast, accurate, nondestructive, and can be performed outdoors in full-sun conditions. We develop an interferometric model that estimates coating porosity, thickness, and fractional area coverage from the measured reflectance spectrum for a uniform single-layer coating. We demonstrate the measurement outdoors on an active PV installation, identify the presence of an ARC, and estimate the properties of the coating.
- Open Access
- 21 January 2021
- Published by Springer Nature in Scientific Reports
- Vol. 11 (1) , 1-13
Towards validation of combined-accelerated stress testing through failure analysis of polyamide-based photovoltaic backsheets
Novel methods for advancing reliability testing of photovoltaic (PV) modules and materials have recently been developed. Combined-accelerated stress testing (C-AST) is one such method which has demonstrated reliable reproduction of some field-failures which were not reproducible by standard certification tests. To increase confidence and assist in the development of C-AST, and other new testing protocols, it is important to validate that the failure modes observed and mechanisms induced are representative of those observed in the field, and not the product of unrealistic stress conditions. Here we outline a method using appropriate materials characterization and modelling to validate the failure mechanisms induced in C-AST such that we can increase confidence in the test protocol. The method is demonstrated by applying it to a known cracking failure of a specific polyamide (PA)-based backsheet material. We found that the failure of the PA-based backsheet was a result of a combination of stress factors. Photo-oxidation from ultra-violet (UV) radiation exposure caused a reduction in fracture toughness, which ultimately lead to the cracking failure. We show that the chemical and structural changes observed in the backsheet following C-AST aging were also observed in field-aged samples. These results increase confidence that the conditions applied in C-AST are representative of the field and demonstrates our approach to validating the failure mechanisms induced.
- 21 January 2021
- Published by Springer Nature in Scientific Reports
- Vol. 11 (1) , 1-13
Open AccessTowards validation of combined-accelerated stress testing through failure analysis of polyamide-based photovoltaic backsheets
Novel methods for advancing reliability testing of photovoltaic (PV) modules and materials have recently been developed. Combined-accelerated stress testing (C-AST) is one such method which has demonstrated reliable reproduction of some field-failures which were not reproducible by standard certification tests. To increase confidence and assist in the development of C-AST, and other new testing protocols, it is important to validate that the failure modes observed and mechanisms induced are representative of those observed in the field, and not the product of unrealistic stress conditions. Here we outline a method using appropriate materials characterization and modelling to validate the failure mechanisms induced in C-AST such that we can increase confidence in the test protocol. The method is demonstrated by applying it to a known cracking failure of a specific polyamide (PA)-based backsheet material. We found that the failure of the PA-based backsheet was a result of a combination of stress factors. Photo-oxidation from ultra-violet (UV) radiation exposure caused a reduction in fracture toughness, which ultimately lead to the cracking failure. We show that the chemical and structural changes observed in the backsheet following C-AST aging were also observed in field-aged samples. These results increase confidence that the conditions applied in C-AST are representative of the field and demonstrates our approach to validating the failure mechanisms induced.
- Open Access
- 17 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Photovoltaics
- Vol. 10 (3) , 888-897
Photovoltaic String Sizing Using Site-Specific Modeling
One key design decision for photovoltaic (PV) power plants is to select the number of PV modules connected in series, also called the string size. Longer strings typically lower total system costs, but the string size must still meet relevant electrical standards to ensure that the maximum system voltage remains less than the design voltage. Traditional methods calculate string size using the temperature coefficient of open-circuit voltage assuming that the coldest expected temperature occurs simultaneously with a full-sun irradiance of ${\text{1000}}\,{\text{W/m}}^{\text{2}}$ . Here, we demonstrate that this traditional method is unnecessarily conservative, resulting in a string size $\sim$ 10% shorter than necessary to maintain system voltage within limits. Instead, engineers may determine string size by modeling open-circuit voltage over time using historical weather data, a method consistent with the 2017 National Electric Code. For bifacial systems, we derive a simple additive term that predicts the additional voltage rise. We demonstrate that this site-specific modeling procedure predicts open-circuit voltages in close agreement with data from field measurements. We further perform a comprehensive sensitivity analysis to identify an appropriate safety factor. By using site-specific modeling instead of traditional methods, we estimate a $\sim$ 1.2% reduction in levelized cost of electricity, a significant improvement to PV power plant economics. The method is provided as an easy-to-use web tool and as an open-source Python package.- 17 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Photovoltaics
- Vol. 10 (3) , 888-897
Open AccessPhotovoltaic String Sizing Using Site-Specific Modeling
One key design decision for photovoltaic (PV) power plants is to select the number of PV modules connected in series, also called the string size. Longer strings typically lower total system costs, but the string size must still meet relevant electrical standards to ensure that the maximum system voltage remains less than the design voltage. Traditional methods calculate string size using the temperature coefficient of open-circuit voltage assuming that the coldest expected temperature occurs simultaneously with a full-sun irradiance of ${\text{1000}}\,{\text{W/m}}^{\text{2}}$ . Here, we demonstrate that this traditional method is unnecessarily conservative, resulting in a string size $\sim$ 10% shorter than necessary to maintain system voltage within limits. Instead, engineers may determine string size by modeling open-circuit voltage over time using historical weather data, a method consistent with the 2017 National Electric Code. For bifacial systems, we derive a simple additive term that predicts the additional voltage rise. We demonstrate that this site-specific modeling procedure predicts open-circuit voltages in close agreement with data from field measurements. We further perform a comprehensive sensitivity analysis to identify an appropriate safety factor. By using site-specific modeling instead of traditional methods, we estimate a $\sim$ 1.2% reduction in levelized cost of electricity, a significant improvement to PV power plant economics. The method is provided as an easy-to-use web tool and as an open-source Python package. - 4 August 2014
- Published by AIP Publishing in Applied Physics Letters
- Vol. 105 (5) , 053106
Alignment of the diamond nitrogen vacancy center by strain engineering
The nitrogen vacancy (NV) center in diamond is a sensitive probe of magnetic field and a promising qubit candidate for quantum information processing. The performance of many NV-based devices improves by aligning the NV(s) parallel to a single crystallographic direction. Using ab initio theoretical techniques, we show that NV orientation can be controlled by high-temperature annealing in the presence of strain under currently accessible experimental conditions. We find that (89 ± 7)% of NVs align along the [111] crystallographic direction under 2% compressive biaxial strain (perpendicular to [111]) and an annealing temperature of 970 °C.
- 4 August 2014
- Published by AIP Publishing in Applied Physics Letters
- Vol. 105 (5) , 053106
Alignment of the diamond nitrogen vacancy center by strain engineering
The nitrogen vacancy (NV) center in diamond is a sensitive probe of magnetic field and a promising qubit candidate for quantum information processing. The performance of many NV-based devices improves by aligning the NV(s) parallel to a single crystallographic direction. Using ab initio theoretical techniques, we show that NV orientation can be controlled by high-temperature annealing in the presence of strain under currently accessible experimental conditions. We find that (89 ± 7)% of NVs align along the [111] crystallographic direction under 2% compressive biaxial strain (perpendicular to [111]) and an annealing temperature of 970 °C.