Comparison of line-wise pl-imaging and area-wise pl-imaging

Höffler, Hannes; Dost, Georg; Brand, Andreas A.; Haunschild, Jonas; Schremmer, Hans; Bergmann, Andreas

doi:10.1016/j.egypro.2017.09.341

Cited by 5 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dataset contains 7300 mc-Si wafers and highperformance multicrystalline silicon (HPMC-Si) [14] wafers with a size of 156 × 156 mm from 74 bricks and 25 boxes of ten different manufacturers. The as-cut wafers were measured with an inline PL system with a line-scan camera with an InGaAs detector, as described by Höffler et al [15] with a resolution of 1024 × 1024 pixels. The samples are processed to passivated emitter and rear cells [16] within an industrial production line.…”

Section: A Datasetmentioning

confidence: 99%

Visualizing Material Quality and Similarity of mc-Si Wafers Learned by Convolutional Regression Networks

Demant

Virtue

Kovvali

et al. 2019

IEEE J. Photovoltaics

View full text Add to dashboard Cite

Convolutional neural networks can be trained to assess the material quality of multicrystalline silicon wafers. A successful rating model has been presented in a related work, which directly evaluates the photoluminescence (PL) image of the wafer to predict the current-voltage parameters after solar cell production. This paper presents the results of two visualization techniques to understand what has been learned in the network. First, we reveal what has been learned in the PL image by visualizing the spatial quality distribution of the wafers based on the activation maps of the network. The method is denoted as regression activation mapping. We compare regression activation maps with j 0 images of solar cells to show the semantically meaningful representation of the trained features. Second, we show what has been learned in the data by mapping the learned network representation of all wafers into a low-dimensional subspace. Visualizations reveal the smoothness of our representation with respect to the PL input and measured quality. This technique can be used to detect material anomalies or process faults for samples with high prediction errors.

show abstract

Section: A Datasetmentioning

confidence: 99%

Visualizing Material Quality and Similarity of mc-Si Wafers Learned by Convolutional Regression Networks

Demant

Virtue

Kovvali

et al. 2019

IEEE J. Photovoltaics

View full text Add to dashboard Cite

show abstract

“…In particular, line‐scan photoluminescence (PL) and contactless electroluminescence (EL) are increasingly employed for module quality inspection, and both techniques rely on subcell illumination patterns rather than full, homogeneous illumination . Line‐scan PL is designed to measure the luminescence signal from cell regions coincident with the laser, making this technique particularly useful for in‐line screening . However, line‐scan PL images represent cell regions under relative charge extraction conditions, where the photogenerated voltage difference drives carriers out of the illuminated region.…”

mentioning

confidence: 99%

Imaging Lateral Drift Kinetics to Understand Causes of Outdoor Degradation in Silicon Heterojunction Photovoltaic Modules

2019

View full text Add to dashboard Cite

Lateral drift currents caused by partial or patterned illumination of photovoltaic cells have important implications for laser‐based screening methods and luminescence imaging of fielded and lab‐stressed modules. This study investigates the kinetics of carrier drift and bulk recombination following patterned illumination in commercial silicon heterojunctions with intrinsic thin layer (HIT) modules, comparing the kinetics between a ten‐year field‐weathered module versus a control module that is stored indoors. The measurement of the microwave photoconductance decay (μPCD) transients in the modules, both coincident with the photoexcitation and in nonilluminated cell regions, reveals carrier drift on the 100 μs timescale, followed by millisecond bulk lifetimes. Importantly, the μPCD transients are consistent with luminescence spreading over the cell, which is imaged using a time‐gated InGaAs array camera. The weathered HIT module shows slower lateral drift and faster bulk lifetimes compared with the control, suggesting increased series resistance and accelerated nonradiative recombination in this module, attributable to degradation of the transparent conductive oxide and degradation of the passivation layer, respectively. These results provide a novel example of using time‐resolved measurements directly on full photovoltaic modules to reveal causes of degradation, providing insight for further development of module imaging and screening capabilities.

show abstract

Comparison of photovoltaic module luminescence imaging techniques: Assessing the influence of lateral currents in high-efficiency device structures

Sulas

Johnston

Jordan

2019

Solar Energy Materials and Solar Cells

View full text Add to dashboard Cite

Comparison of line-wise pl-imaging and area-wise pl-imaging

Cited by 5 publications

References 20 publications

Visualizing Material Quality and Similarity of mc-Si Wafers Learned by Convolutional Regression Networks

Visualizing Material Quality and Similarity of mc-Si Wafers Learned by Convolutional Regression Networks

Imaging Lateral Drift Kinetics to Understand Causes of Outdoor Degradation in Silicon Heterojunction Photovoltaic Modules

Comparison of photovoltaic module luminescence imaging techniques: Assessing the influence of lateral currents in high-efficiency device structures

Contact Info

Product

Resources

About