Electroluminescent Image Processing and Cell Degradation Type Classification via Computer Vision and Statistical Learning Methodologies

Fada, Justin S.; Hossain, Mohammad A.; Braid, Jennifer L.; Yang, Shuying; Peshek, Timothy J.; French, Roger H.

doi:10.1109/pvsc.2017.8366291

Cited by 15 publications

(8 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To overcome these limitations, this study proposes a machine learning (ML) based approach to extract the defect parameters from lifetime curves. ML-based methods are already used at the PV system level, for example for fault detection 23,24 or to identify cracks in modules using luminescence imaging techniques [25][26][27][28] . ML has also been used in non-Si applications to find relevant material parameters for fabrication of CIGS solar cells 29 , multijunction solar cells 30 , organic solar cells 31 , or perovskite solar cells 32 .…”

Section: Introductionmentioning

confidence: 99%

Extracting bulk defect parameters in silicon wafers using machine learning models

et al. 2020

View full text Add to dashboard Cite

The performance of high-efficiency silicon solar cells is limited by the presence of bulk defects. Identification of these defects has the potential to improve cell performance and reliability. The impact of bulk defects on minority carrier lifetime is commonly measured using temperature- and injection-dependent lifetime spectroscopy and the defect parameters, such as its energy level and capture cross-section ratio, are usually extracted by fitting the Shockley-Read-Hall equation. We propose an alternative extraction approach by using machine learning trained on more than a million simulated lifetime curves, achieving coefficient of determinations between the true and predicted values of the defect parameters above 99%. In particular, random forest regressors, show that defect energy levels can be predicted with a high precision of ±0.02 eV, 87% of the time. The traditional approach of fitting to the Shockley-Read-Hall equation usually yields two sets of defect parameters, one in each half bandgap. The machine learning model is trained to predict the half bandgap location of the energy level, and successfully overcome the traditional approach’s limitation. The proposed approach is validated using experimental measurements, where the machine learning predicts defect energy level and capture cross-section ratio within the uncertainty range of the traditional fitting method. The successful application of machine learning in the context of bulk defect parameter extraction paves the way to more complex data-driven physical models which have the potential to overcome the limitation of traditional approaches and can be applied to other materials such as perovskite and thin film.

show abstract

Section: Introductionmentioning

confidence: 99%

Extracting bulk defect parameters in silicon wafers using machine learning models

et al. 2020

View full text Add to dashboard Cite

show abstract

“…In the case of PV panels, one approach is to identify corners in the acquired image and rectify the image, so those corners are mapped to a rectangle or a square depending on the characteristics of the PV panel [12], and an implementation requiring manual marking of the panel corners can be found [13]. An automatic method was designed in [14] for "small angle rotation" and panels containing "moderate degradation of the outer cell." As the authors assume that the panel is the main structure present in the image, their method cannot generalize to outdoor daylight acquisition.…”

Section: Problem Formulation and Previous Workmentioning

confidence: 99%

Method for Estimation and Correction of Perspective Distortion of Electroluminescence Images of Photovoltaic Panels

Mantel

Villebro

Parikh

et al. 2020

IEEE J. Photovoltaics

View full text Add to dashboard Cite

The number of photovoltaic panels installed globally is continuously growing, requiring an automatic inspection procedure for operation and maintenance. Drones can be a useful tool to this aim as they enable fast acquisition of various imaging modalities: visual, infrared, or electroluminescence (EL). Image distortions due to perspective must be corrected to allow further automatic processing. It can be done by estimating the corresponding rotation angles to control the camera gimbal or as postprocessing to rectify the images. This article presents two methods to achieve both goals by identifying known points in the acquired image. The first method detects the four panel corners, whereas the second method finds the corners of each cell. The performance evaluation is performed first quantitatively on a validation dataset composed of 113 EL images and their corresponding ground-truth orientations. A qualitative evaluation shows satisfying performance of the rectification similarly for both methods. The quantitative performance is varying for each rotation axis. The average absolute error is 2.78 • along the x-axis, 2.64 • along the y-axis, and 1.28 • along the z-axis for the panel method and 3.26 • , 2.05 • , and 1.24 • for the cell method. As a proof of concept, a final test on drone-acquired EL images shows good performance for the image rectification in real-life conditions.

show abstract

“…Analyzing EL images is typically time-consuming [14] and requires expert knowledge regarding the different defects. It is, therefore, expensive to perform on a large scale [15]. One possible path to improve the analysis is using machine learning (ML) to detect different defects more accurately.…”

Section: Introductionmentioning

confidence: 99%

Solar Cell Cracks and Finger Failure Detection Using Statistical Parameters of Electroluminescence Images and Machine Learning

et al. 2020

View full text Add to dashboard Cite

A wide range of defects, failures, and degradation can develop at different stages in the lifetime of photovoltaic modules. To accurately assess their effect on the module performance, these failures need to be quantified. Electroluminescence (EL) imaging is a powerful diagnostic method, providing high spatial resolution images of solar cells and modules. EL images allow the identification and quantification of different types of failures, including those in high recombination regions, as well as series resistance-related problems. In this study, almost 46,000 EL cell images are extracted from photovoltaic modules with different defects. We present a method that extracts statistical parameters from the histogram of these images and utilizes them as a feature descriptor. Machine learning algorithms are then trained using this descriptor to classify the detected defects into three categories: (i) cracks (Mode B and C), (ii) micro-cracks (Mode A) and finger failures, and (iii) no failures. By comparing the developed methods with the commonly used one, this study demonstrates that the pre-processing of images into a feature vector of statistical parameters provides a higher classification accuracy than would be obtained by raw images alone. The proposed method can autonomously detect cracks and finger failures, enabling outdoor EL inspection using a drone-mounted system for quick assessments of photovoltaic fields.

show abstract

Electroluminescent Image Processing and Cell Degradation Type Classification via Computer Vision and Statistical Learning Methodologies

Cited by 15 publications

References 17 publications

Extracting bulk defect parameters in silicon wafers using machine learning models

Extracting bulk defect parameters in silicon wafers using machine learning models

Method for Estimation and Correction of Perspective Distortion of Electroluminescence Images of Photovoltaic Panels

Solar Cell Cracks and Finger Failure Detection Using Statistical Parameters of Electroluminescence Images and Machine Learning

Contact Info

Product

Resources

About