Detection of microcracks and dark spots in monocrystalline PERC cells using photoluminescene imaging and YOLO-based CNN with spatial pyramid pooling
Two common defects encountered during manufacturing of crystalline silicon solar cells are microcrack and dark spot or dark region. The microcrack in particular is a major threat to module performance since it is responsible for most PV failures and other types of damage in the field. On the other h...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
EDP Sciences
2022-01-01
|
| Series: | EPJ Photovoltaics |
| Subjects: | |
| Online Access: | https://www.epj-pv.org/articles/epjpv/full_html/2022/01/pv220063/pv220063.html |
| _version_ | 1811301513239199744 |
|---|---|
| author | Binomairah Amran Abdullah Azizi Khoo Bee Ee Mahdavipour Zeinab Teo Teow Wee Mohd Noor Nor Shahirah Abdullah Mohd Zaid |
| author_facet | Binomairah Amran Abdullah Azizi Khoo Bee Ee Mahdavipour Zeinab Teo Teow Wee Mohd Noor Nor Shahirah Abdullah Mohd Zaid |
| author_sort | Binomairah Amran |
| collection | DOAJ |
| description | Two common defects encountered during manufacturing of crystalline silicon solar cells are microcrack and dark spot or dark region. The microcrack in particular is a major threat to module performance since it is responsible for most PV failures and other types of damage in the field. On the other hand, dark region in which one cell or part of the cell appears darker under UV illumination is mainly responsible for PV reduced efficiency, and eventually lost of performance. Therefore, one key challenge for solar cell manufacturers is to remove defective cells from further processing. Recently, few researchers have investigated deep learning as an alternative approach for defect detection in solar cell manufacturing. The results are quite encouraging. This paper evaluates the convolutional neural network based on heavy-weighted You Only Look Once (YOLO) version 4 or YOLOv4 and the tiny version of this algorithm referred here as Tiny-YOLOv4. Experimental results suggest that the multi-class YOLOv4 is the best model in term of mean average precision (mAP) and prediction time, averaging at 98.8% and 62.9 ms respectively. Meanwhile an improved Tiny-YOLOv4 with Spatial Pyramid Pooling scheme resulted in mAP of 91.0% and runtime of 28.2 ms. Even though the tiny-weighted YOLOv4 performs slightly lower compared to its heavy-weighted counterpart, however the runtime of the former is 2.2 order much faster than the later. |
| first_indexed | 2024-04-13T07:10:12Z |
| format | Article |
| id | doaj.art-75d1cb3721ab4ddfa3771c00159594c4 |
| institution | Directory Open Access Journal |
| issn | 2105-0716 |
| language | English |
| last_indexed | 2024-04-13T07:10:12Z |
| publishDate | 2022-01-01 |
| publisher | EDP Sciences |
| record_format | Article |
| series | EPJ Photovoltaics |
| spelling | doaj.art-75d1cb3721ab4ddfa3771c00159594c42022-12-22T02:56:53ZengEDP SciencesEPJ Photovoltaics2105-07162022-01-01132710.1051/epjpv/2022025pv220063Detection of microcracks and dark spots in monocrystalline PERC cells using photoluminescene imaging and YOLO-based CNN with spatial pyramid poolingBinomairah Amran0Abdullah Azizi1Khoo Bee Ee2Mahdavipour Zeinab3Teo Teow Wee4Mohd Noor Nor Shahirah5Abdullah Mohd Zaid6School of Electrical and Electronics Engineering, Engineering Campus, Universiti Sains MalaysiaFaculty of Information Sciences and Technology, Universiti Kebangsaan MalaysiaSchool of Electrical and Electronics Engineering, Engineering Campus, Universiti Sains MalaysiaTT Vision Technologies Sdn. Bhd., Plot 106, Hilir Sungai Keluang 5, Bayan Lepas Industrial Zone, Phase 4TT Vision Technologies Sdn. Bhd., Plot 106, Hilir Sungai Keluang 5, Bayan Lepas Industrial Zone, Phase 4School of Electrical and Electronics Engineering, Engineering Campus, Universiti Sains MalaysiaSchool of Electrical and Electronics Engineering, Engineering Campus, Universiti Sains MalaysiaTwo common defects encountered during manufacturing of crystalline silicon solar cells are microcrack and dark spot or dark region. The microcrack in particular is a major threat to module performance since it is responsible for most PV failures and other types of damage in the field. On the other hand, dark region in which one cell or part of the cell appears darker under UV illumination is mainly responsible for PV reduced efficiency, and eventually lost of performance. Therefore, one key challenge for solar cell manufacturers is to remove defective cells from further processing. Recently, few researchers have investigated deep learning as an alternative approach for defect detection in solar cell manufacturing. The results are quite encouraging. This paper evaluates the convolutional neural network based on heavy-weighted You Only Look Once (YOLO) version 4 or YOLOv4 and the tiny version of this algorithm referred here as Tiny-YOLOv4. Experimental results suggest that the multi-class YOLOv4 is the best model in term of mean average precision (mAP) and prediction time, averaging at 98.8% and 62.9 ms respectively. Meanwhile an improved Tiny-YOLOv4 with Spatial Pyramid Pooling scheme resulted in mAP of 91.0% and runtime of 28.2 ms. Even though the tiny-weighted YOLOv4 performs slightly lower compared to its heavy-weighted counterpart, however the runtime of the former is 2.2 order much faster than the later.https://www.epj-pv.org/articles/epjpv/full_html/2022/01/pv220063/pv220063.htmlsolar cellmicrocrackdark regioncnnyolo |
| spellingShingle | Binomairah Amran Abdullah Azizi Khoo Bee Ee Mahdavipour Zeinab Teo Teow Wee Mohd Noor Nor Shahirah Abdullah Mohd Zaid Detection of microcracks and dark spots in monocrystalline PERC cells using photoluminescene imaging and YOLO-based CNN with spatial pyramid pooling EPJ Photovoltaics solar cell microcrack dark region cnn yolo |
| title | Detection of microcracks and dark spots in monocrystalline PERC cells using photoluminescene imaging and YOLO-based CNN with spatial pyramid pooling |
| title_full | Detection of microcracks and dark spots in monocrystalline PERC cells using photoluminescene imaging and YOLO-based CNN with spatial pyramid pooling |
| title_fullStr | Detection of microcracks and dark spots in monocrystalline PERC cells using photoluminescene imaging and YOLO-based CNN with spatial pyramid pooling |
| title_full_unstemmed | Detection of microcracks and dark spots in monocrystalline PERC cells using photoluminescene imaging and YOLO-based CNN with spatial pyramid pooling |
| title_short | Detection of microcracks and dark spots in monocrystalline PERC cells using photoluminescene imaging and YOLO-based CNN with spatial pyramid pooling |
| title_sort | detection of microcracks and dark spots in monocrystalline perc cells using photoluminescene imaging and yolo based cnn with spatial pyramid pooling |
| topic | solar cell microcrack dark region cnn yolo |
| url | https://www.epj-pv.org/articles/epjpv/full_html/2022/01/pv220063/pv220063.html |
| work_keys_str_mv | AT binomairahamran detectionofmicrocracksanddarkspotsinmonocrystallineperccellsusingphotoluminesceneimagingandyolobasedcnnwithspatialpyramidpooling AT abdullahazizi detectionofmicrocracksanddarkspotsinmonocrystallineperccellsusingphotoluminesceneimagingandyolobasedcnnwithspatialpyramidpooling AT khoobeeee detectionofmicrocracksanddarkspotsinmonocrystallineperccellsusingphotoluminesceneimagingandyolobasedcnnwithspatialpyramidpooling AT mahdavipourzeinab detectionofmicrocracksanddarkspotsinmonocrystallineperccellsusingphotoluminesceneimagingandyolobasedcnnwithspatialpyramidpooling AT teoteowwee detectionofmicrocracksanddarkspotsinmonocrystallineperccellsusingphotoluminesceneimagingandyolobasedcnnwithspatialpyramidpooling AT mohdnoornorshahirah detectionofmicrocracksanddarkspotsinmonocrystallineperccellsusingphotoluminesceneimagingandyolobasedcnnwithspatialpyramidpooling AT abdullahmohdzaid detectionofmicrocracksanddarkspotsinmonocrystallineperccellsusingphotoluminesceneimagingandyolobasedcnnwithspatialpyramidpooling |