Degradation Mechanism Detection in Photovoltaic Backsheets by Fully Convolutional Neural Network

Zhang, Binbin; Grant, Joydan; Bruckman, Laura S.; Wodo, Olga; Rai, Rahul

doi:10.1038/s41598-019-52550-6

Title: Degradation Mechanism Detection in Photovoltaic Backsheets by Fully Convolutional Neural Network

Journal Article · Wed Nov 06 00:00:00 EST 2019 · Scientific Reports

DOI:https://doi.org/10.1038/s41598-019-52550-6· OSTI ID:1799274

Zhang, Binbin ^[1]; Grant, Joydan ^[2];

^[3]; Wodo, Olga ^[4]; Rai, Rahul ^[1]

Univ. at Buffalo, NY (United States). Mechanical and Aerospace Engineering Dept.
Mechanical Engineering Tuskegee Univ., Tuskegee, AL (United States)
Case Western Reserve Univ., Cleveland, OH (United States). Dept. of Materials Science and Engineering
Univ. at Buffalo, NY (United States). Mechanical and Aerospace Engineering Dept.; Univ. at Buffalo, NY (United States). Materials Design and Innovation Dept.

Materials and devices age with time. Material aging and degradation has important implications for lifetime performance of materials and systems. While consensus exists that materials should be studied and designed for degradation, materials inspection during operation is typically performed manually by technicians. The manual inspection makes studies prone to errors and uncertainties due to human subjectivity. In this work, we focus on automating the process of degradation mechanism detection through the use of a fully convolutional deep neural network architecture (F-CNN). We demonstrate that F-CNN architecture allows for automated inspection of cracks in polymer backsheets from photovoltaic (PV) modules. The developed F-CNN architecture enabled an end-to-end semantic inspection of the PV module backsheets by applying a contracting path of convolutional blocks (encoders) followed by an expansive path of decoding blocks (decoders). First, the hierarchy of contextual features is learned from the input images by encoders. Next, these features are reconstructed to the pixel-level prediction of the input by decoders. The structure of the encoder and the decoder networks are thoroughly investigated for the multi-class pixel-level degradation type prediction for PV module backsheets. The developed F-CNN framework is validated by reporting degradation type prediction accuracy for the pixel level prediction at the level of 92.8%.

View Accepted Manuscript (DOE)

Cite

Export

Save

Research Organization:: UL, LLC, Northbrook, IL (United States)

Sponsoring Organization:: USDOE Office of Energy Efficiency and Renewable Energy (EERE)

Grant/Contract Number:: EE0007143

OSTI ID:: 1799274

Journal Information:: Scientific Reports, Vol. 9, Issue 1; ISSN 2045-2322

Publisher:: Nature Publishing GroupCopyright Statement

Country of Publication:: United States

Language:: English

References (43)

Moisture transport, adhesion, and corrosion protection of PV module packaging materials Jorgensen, G.; Terwilliger, K.; Delcueto, J. Solar Energy Materials and Solar Cells, Vol. 90, Issue 16 https://doi.org/10.1016/j.solmat.2006.04.003	journal	October 2006
A Visual Detection System for Rail Surface Defects Li, Qingyong; Ren, Shengwei IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), Vol. 42, Issue 6 https://doi.org/10.1109/TSMCC.2012.2198814	journal	November 2012
Multivariate multiple regression models of poly(ethylene-terephthalate) film degradation under outdoor and multi-stressor accelerated weathering exposures Gordon, Devin A.; Huang, Wei-Heng; Burns, David M. PLOS ONE, Vol. 13, Issue 12 https://doi.org/10.1371/journal.pone.0209016	journal	December 2018
Learning Hierarchical Features for Scene Labeling Farabet, Clement; Couprie, Camille; Najman, Laurent IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 35, Issue 8 https://doi.org/10.1109/TPAMI.2012.231	journal	August 2013
Fully Convolutional Networks for Semantic Segmentation Shelhamer, Evan; Long, Jonathan; Darrell, Trevor arXiv https://doi.org/10.48550/arxiv.1605.06211	preprint	January 2016
LabelMe: A Database and Web-Based Tool for Image Annotation Russell, Bryan C.; Torralba, Antonio; Murphy, Kevin P. International Journal of Computer Vision, Vol. 77, Issue 1-3 https://doi.org/10.1007/s11263-007-0090-8	journal	October 2007
Learning from Imbalanced Data IEEE Transactions on Knowledge and Data Engineering, Vol. 21, Issue 9 https://doi.org/10.1109/TKDE.2008.239	journal	September 2009
Steel defect classification with Max-Pooling Convolutional Neural Networks Masci, Jonathan; Meier, Ueli; Ciresan, Dan 2012 International Joint Conference on Neural Networks (IJCNN 2012 - Brisbane), The 2012 International Joint Conference on Neural Networks (IJCNN) https://doi.org/10.1109/IJCNN.2012.6252468	conference	June 2012
On-Site Electroluminescence Study of Field-Aged PV Modules Dubey, Rajiv; Vasi, Juzer; Chattopadhyay, Shashwata 2018 IEEE 7th World Conference on Photovoltaic Energy Conversion (WCPEC) (A Joint Conference of 45th IEEE PVSC, 28th PVSEC & 34th EU PVSEC) https://doi.org/10.1109/PVSC.2018.8548080	conference	June 2018
Outdoor photoluminescence imaging of photovoltaic modules with sunlight excitation Bhoopathy, Raghavi; Kunz, Oliver; Juhl, Mattias Progress in Photovoltaics: Research and Applications, Vol. 26, Issue 1 https://doi.org/10.1002/pip.2946	journal	September 2017
Aging mechanisms of polymeric films for PV encapsulation Oreski, G.; Wallner, G. M. Solar Energy, Vol. 79, Issue 6 https://doi.org/10.1016/j.solener.2005.02.008	journal	December 2005
Delamination behaviour of multi-layer films for PV encapsulation Oreski, G.; Wallner, G. Solar Energy Materials and Solar Cells, Vol. 89, Issue 2-3 https://doi.org/10.1016/j.solmat.2005.02.009	journal	November 2005
Analysis of degradation mechanisms of crystalline silicon PV modules after 12 years of operation in Southern Europe Sánchez-Friera, Paula; Piliougine, Michel; Peláez, Javier Progress in Photovoltaics: Research and Applications, Vol. 19, Issue 6 https://doi.org/10.1002/pip.1083	journal	January 2011
The Pascal Visual Object Classes Challenge: A Retrospective Everingham, Mark; Eslami, S. M. Ali; Van Gool, Luc International Journal of Computer Vision, Vol. 111, Issue 1 https://doi.org/10.1007/s11263-014-0733-5	journal	June 2014
Utility-scale solar PV performance enhancements through system-level modifications Glick, Andrew; Ali, Naseem; Bossuyt, Juliaan Scientific Reports, Vol. 10, Issue 1 https://doi.org/10.1038/s41598-020-66347-5	journal	June 2020
Backsheet and Module Durability and Performance and Comparison of Accelerated Testing to Long Term Fielded Modules Gambogi, W. J.; Kopchick, J. G.; Felder, T. C. WIP https://doi.org/10.4229/28theupvsec2013-4co.10.2	text	January 2013
Fully Convolutional Networks for Semantic Segmentation Shelhamer, Evan; Long, Jonathan; Darrell, Trevor IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 39, Issue 4 https://doi.org/10.1109/tpami.2016.2572683	journal	April 2017
Commonly observed degradation in field-aged photovoltaic modules Quintana, M. A.; King, D. L.; McMahon, T. J. Conference Record of the Twenty-Ninth IEEE Photovoltaic Specialists Conference 2002, Conference Record of the Twenty-Ninth IEEE Photovoltaic Specialists Conference, 2002. https://doi.org/10.1109/PVSC.2002.1190879	conference	January 2002
Classification of steel microstructures using Modified Alternate Local Ternary Pattern Arivazhagan, S.; Tracia, J. Jasline; Selvakumar, N. Materials Research Express, Vol. 6, Issue 9 https://doi.org/10.1088/2053-1591/ab2d83	journal	July 2019
Deconvolutional networks Zeiler, Matthew D.; Krishnan, Dilip; Taylor, Graham W. 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition https://doi.org/10.1109/CVPR.2010.5539957	conference	June 2010
Crack Identification in beam Structures Using Mechanical Impedance Bamnios, Y.; Douka, E.; Trochidis, A. Journal of Sound and Vibration, Vol. 256, Issue 2 https://doi.org/10.1006/jsvi.2001.4209	journal	September 2002
A review on computer vision based defect detection and condition assessment of concrete and asphalt civil infrastructure Koch, Christian; Georgieva, Kristina; Kasireddy, Varun Advanced Engineering Informatics, Vol. 29, Issue 2 https://doi.org/10.1016/j.aei.2015.01.008	journal	April 2015
Fully convolutional networks for semantic segmentation Long, Jonathan; Shelhamer, Evan; Darrell, Trevor 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2015.7298965	conference	June 2015
Microstructural diagram for steel based on crystallography with machine learning Tsutsui, Kazumasa; Terasaki, Hidenori; Maemura, Tatsuya Computational Materials Science, Vol. 159 https://doi.org/10.1016/j.commatsci.2018.12.003	journal	March 2019
U-Net: Convolutional Networks for Biomedical Image Segmentation Ronneberger, Olaf; Fischer, Philipp; Brox, Thomas Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III https://doi.org/10.1007/978-3-319-24574-4_28	book	November 2015
Hypercolumns for object segmentation and fine-grained localization Hariharan, Bharath; Arbelaez, Pablo; Girshick, Ross 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2015.7298642	conference	June 2015
Large-area, high-resolution characterisation and classification of damage mechanisms in dual-phase steel using deep learning Kusche, Carl; Reclik, Tom; Freund, Martina PLOS ONE, Vol. 14, Issue 5 https://doi.org/10.1371/journal.pone.0216493	journal	May 2019
Learning Deconvolution Network for Semantic Segmentation Noh, Hyeonwoo; Hong, Seunghoon; Han, Bohyung 2015 IEEE International Conference on Computer Vision (ICCV) https://doi.org/10.1109/ICCV.2015.178	conference	December 2015
Large-area, high-resolution characterisation and classification of damage mechanisms in dual-phase steel using deep learning Kusche, Carl; Reclik, Tom; Freund, Martina RWTH Aachen University https://doi.org/10.18154/rwth-2019-10367	text	January 2019
Feature selection, L 1 vs. L 2 regularization, and rotational invariance Ng, Andrew Y. Twenty-first international conference on Machine learning - ICML '04 https://doi.org/10.1145/1015330.1015435	conference	January 2004
Visual Degradation in Field-Aged Crystalline Silicon PV Modules in India and Correlation With Electrical Degradation Chattopadhyay, Shashwata; Dubey, Rajiv; Kuthanazhi, Vivek IEEE Journal of Photovoltaics, Vol. 4, Issue 6 https://doi.org/10.1109/JPHOTOV.2014.2356717	journal	November 2014
Fully Convolutional Architectures for Multiclass Segmentation in Chest Radiographs Novikov, Alexey A.; Lenis, Dimitrios; Major, David IEEE Transactions on Medical Imaging, Vol. 37, Issue 8 https://doi.org/10.1109/TMI.2018.2806086	journal	August 2018
High through-put, outdoor characterization of silicon photovoltaic modules by moving electroluminescence measurements Doll, Bernd; Pickel, Tobias; Schreer, Oliver Infrared Sensors, Devices, and Applications VIII https://doi.org/10.1117/12.2320518	conference	September 2018
Automatic surface defect detection for mobile phone screen glass based on machine vision Jian, Chuanxia; Gao, Jian; Ao, Yinhui Applied Soft Computing, Vol. 52 https://doi.org/10.1016/j.asoc.2016.10.030	journal	March 2017
A Comparison of Key PV Backsheet and Module Performance from Fielded Module Exposures and Accelerated Tests Gambogi, W.; Heta, Y.; Hashimoto, K. IEEE Journal of Photovoltaics, Vol. 4, Issue 3 https://doi.org/10.1109/JPHOTOV.2014.2305472	journal	May 2014
Learning Deconvolution Network for Semantic Segmentation Noh, Hyeonwoo; Hong, Seunghoon; Han, Bohyung arXiv https://doi.org/10.48550/arxiv.1505.04366	preprint	January 2015
Technology advances needed for photovoltaics to achieve widespread grid price parity: Widespread grid price parity for photovoltaics Jones-Albertus, Rebecca; Feldman, David; Fu, Ran Progress in Photovoltaics: Research and Applications, Vol. 24, Issue 9 https://doi.org/10.1002/pip.2755	journal	April 2016
A non-destructive method for crack quantification in photovoltaic backsheets under accelerated and real-world exposures Klinke, Addison G.; Gok, Abdulkerim; Ifeanyi, Silas I. Polymer Degradation and Stability, Vol. 153 https://doi.org/10.1016/j.polymdegradstab.2018.05.008	journal	July 2018
Automatic ultrasonic inspection for internal defect detection in composite materials D’Orazio, T.; Leo, M.; Distante, A. NDT & E International, Vol. 41, Issue 2 https://doi.org/10.1016/j.ndteint.2007.08.001	journal	March 2008
Temporal evolution and pathway models of poly(ethylene-terephthalate) degradation under multi-factor accelerated weathering exposures Gok, Abdulkerim; Fagerholm, Cara L.; French, Roger H. PLOS ONE, Vol. 14, Issue 2 https://doi.org/10.1371/journal.pone.0212258	journal	February 2019
What is a good evaluation measure for semantic segmentation? Csurka, Gabriela; Larlus, Diane; Perronnin, Florent Procedings of the British Machine Vision Conference 2013 https://doi.org/10.5244/C.27.32	conference	January 2013
Crack identification in beams using wavelet analysis Douka, E.; Loutridis, S.; Trochidis, A. International Journal of Solids and Structures, Vol. 40, Issue 13-14 https://doi.org/10.1016/s0020-7683(03)00147-1	journal	June 2003
Statistical and Domain Analytics Applied to PV Module Lifetime and Degradation Science Bruckman, L. S.; Wheeler, N. R.; Ma, J. IEEE Access, Vol. 1 https://doi.org/10.1109/access.2013.2267611	journal	January 2013

Similar Records

Generalized Spatio-Temporal Model of Backsheet Degradation From Field Surveys of Photovoltaic Modules

Journal Article · Fri Aug 02 00:00:00 EDT 2019 · IEEE Journal of Photovoltaics · OSTI ID:1799274

Wang, Yu; Huang, Wei-Heng; Fairbrother, Andrew; +14 more

Field retrieved photovoltaic backsheet survey from diverse climate zones: Analysis of degradation patterns and phenomena

Journal Article · Wed May 17 00:00:00 EDT 2023 · Solar Energy · OSTI ID:1799274

Wieser, Raymond J.; Wang, Yu; Fairbrother, Andrew; +10 more

Thermally Conductive Backsheets (TCB) for Reduced Operating Temperatures

Technical Report · Tue Aug 20 00:00:00 EDT 2019 · OSTI ID:1799274

TamizhMani, GovindaSamy; Pavgi, Ashwini; Oh, Jaewon; +4 more

Related Subjects

14 SOLAR ENERGY
Science & Technology - Other Topics
Computational methods
Computer science

Title: Degradation Mechanism Detection in Photovoltaic Backsheets by Fully Convolutional Neural Network

Citation Formats

References (43)

Similar Records

Related Subjects