One important aspect of solar energy generation especially in inter-tropical sites is the local variability of clouds. Satellite images do not have temporal resolution enough to nowcast its impacts on solar plants, this monitoring is made by local cameras. However, cloud detection and monitoring are not trivial due to cloud shape dynamics, the camera is a linear and self-adjusting device, with fish-eye lenses generating a flat image that distorts images near the horizon. The present work focuses on cloud identification to predict its effects on solar plants that are distinct for every site’s climatology and geography. We used RASPBERY-PI-based cameras pointed at the horizon to allow observation of clouds’ vertical distribution, not possible with a unique fish-eye lens. A large number of cloud image identification analyses led the researchers to use deep learning methods such as U-net, HRnet, and Detectron. We use transfer learning with weights trained over the “2012 ILSVRC ImageNet” data set and architecture configurations like Resnet, Efficient, and Detectron2. While cloud identification proved a difficult task, we achieved the best results by using Jaccard Coefficient as a validation metric, with the best model being a U-net with Resnet18 using 486？× 648 resolution. This model had an average IoU of 0.6, indicating a satisfactory performance in cloud segmentation. We also observed that the data imbalance affected the overall performance of all models, with the tree class creating a favorable bias. The HRNet model, which works with different resolutions, showed promising results with a more refined segmentation at the pixel level, but it was not necessary to detect the most predominant clouds in the sky. We are currently working on balancing the dataset and mapping out data augmentation transformations for our next experiments. Our ultimate goal is to use such models to predict cloud motion and forecast the impact it will have on solar power generation. The present work has contributed to a better understanding of what techniques work best for cloud identification and paves the way for future studies on the development of a better overall cloud classification model.
References
[1]
Alvares, C. A., Stape, J. L., Sentelhas, P. C., de Moraes Gonçalves, J. L., & Sparovek, G. (2013). Köppen’s Climate Classification Map for Brazil. Meteorologische Zeitschrift, 22, 711-728. https://doi.org/10.1127/0941-2948/2013/0507
[2]
Anagnostos, D., Schmidt, T., Cavadias, S., Soudris, D., Poortmans, J., & Catthoor, F. (2019). A Method for Detailed, Short-Term Energy Yield Forecasting of Photovoltaic Installations. Renewable Energy, 130, 122-129. http://www.sciencedirect.com/science/article/pii/S0960148118307109
[3]
Barrett, E., & Grant, C. K. (1976). The Identification of Cloud Types in Landsat MSS Images. Tech. Rep.
[4]
do Nascimento, L. R., Braga, M., Campos, R. A., Naspolini, H. F., & Rüther, R. (2020). Performance Assessment of Solar Photovoltaic Technologies under Different Climatic Conditions in Brazil. Renewable Energy, 146, 1070-1082. https://www.sciencedirect.com/science/article/pii/S0960148119310006
[5]
do Nascimento, L. R., de Souza Viana, T., Campos, R. A., & Rüther, R. (2019). Extreme Solar Overirradiance Events: Occurrence and Impacts on Utility-Scale Photovoltaic Power Plants in Brazil. Solar Energy, 186, 370-381. https://www.sciencedirect.com/science/article/pii/S0038092X19304530
[6]
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T. et al. (2021). An Image Is Worth 16x16 Words: Transformers for Image Recognition at Scale. http://arxiv.org/abs/2010.11929
[7]
Dubreuil, V., Fante, K., Planchon, O., & Santa’Anna Neto, J. (2018). The Types of Annual Climates in Brazil: An Application of the Classification of Köppen from 1961 to 2015. EchoGéo. https://doi.org/10.4000/echogeo.15017
[8]
Fabel, Y., Nouri, B., Wilbert, S., Blum, N., Triebel, R., Hasenbalg, M. et al. (2022). Applying Self-Supervised Learning for Semantic Cloud Segmentation of All-Sky Images. Atmospheric Measurement Techniques, 15, 797-809. https://amt.copernicus.org/articles/15/797/2022/
[9]
Haeffelin, M., Barthès, L., Bock, O., Boitel, C., Bony, S., Bouniol, D., et al. (2005). SIRTA, A Ground-Based Atmospheric Observatory for Cloud and Aerosol Research. Annales Geophysicae, 23, 253-275. https://angeo.copernicus.org/articles/23/253/2005/angeo-23-253-2005.html
[10]
Hu, Y., & Stamnes, K. (2000). Climate Sensitivity to Cloud Optical Properties. Tellus B, 52, 81-93. https://doi.org/10.3402/tellusb.v52i1.16084
[11]
Juncklaus Martins, B., Cerentini, A., Neto, S. M., & von Wangenheim, A. (2021). Systematic Literature Review on Forecasting/Nowcasting Based upon Ground-Based Cloud Imaging. https://www.researchgate.net/publication/349536859_Systematic_Literature_Review_on_ForecastingNowcasting_based_upon_Ground-Based_Cloud_Imaging
[12]
Juncklaus Martins, B., Cerentini, A., Neto, S. M., & von Wangenheim, A. (2022a). Systematic Review of Nowcasting Approaches for Solar Energy Production Based upon Ground-Based Cloud Imaging. Solar Energy Advances, 2, Article 100019. https://doi.org/10.1016/j.seja.2022.100019
[13]
Juncklaus Martins, B., Polli, M., Cerentini, A., Mantelli, S., Chaves, T., Moreira Branco, N. et al. (2022b). Clouds-1000. Mendeley Data. https://data.mendeley.com/datasets/4pw8vfsnpx/1
[14]
Kumari, P., & Toshniwal, D. (2021). Deep Learning Models for Solar Irradiance Forecasting: A Comprehensive Review. Journal of Cleaner Production, 318, Article 128566. https://www.sciencedirect.com/science/article/pii/S0959652621027736
[15]
Li, P., Dong, L., Xiao, H., & Xu, M. (2015). A Cloud Image Detection Method Based on SVM Vector Machine. Neurocomputing, 169, 34-42. http://www.sciencedirect.com/science/article/pii/S0925231215006864
[16]
Lin, T., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D. et al. (2014). Microsoft Coco: Common Objects in Context. In D. Fleet, T. Pajdla, B. Schiele, & T. Tuytelaars (Eds.), European conference on Computer Vision (pp. 740-755). Springer. https://doi.org/10.1007/978-3-319-10602-1_48
[17]
Long, C. N., Sabburg, J. M., Calbo, J., & Page, J. D. (2006). Retrieving Cloud Characteristics from Ground-Based Daytime Color All-Sky Images. Journal of Atmospheric and Oceanic Technology, 23, 633-652. https://doi.org/10.1175/JTECH1875.1
[18]
Mantelli, S. L., von Wangenheim, A., Pereira, E. B., & Sobieranki, A. C. (2020). Hierarchical Color Similarity Metrics for Step-Wise Application on Sky Monitoring Surface Cameras. Earth and Space Science Open Archive. https://doi.org/10.1002/essoar.10503135.1
[19]
Mantelli, S. L., von Wangenhein, A., Pereira, E. B., & Comunello, E. (2010). The Use of Euclidean Geometric Distance on RGB Color Space for Classification of Sky and Cloud Patterns. Journal of Atmospheric and Oceanic Technology, 27, 1504-1517. https://doi.org/10.1175/2010JTECHA1353.1
[20]
Martins, G. L., Mantelli, S. L., & Rüther, R. (2022). Evaluating the Performance of Radiometers for Solar Overirradiance Events. Solar Energy, 231, 47-56. https://www.sciencedirect.com/science/article/pii/S0038092X21010100
[21]
Mejia, F. A., Kurtz, B., Murray, K., Hinkelman, L. M., Sengupta, M., Xie, Y., & Kleissl, J. (2016). Coupling Sky Images with Radiative Transfer Models: A New Method to Estimate Cloud Optical Depth. Atmospheric Measurement Techniques, 9, 4151-4165. https://www.atmos-meas-tech.net/9/4151/2016/
[22]
Mellit, A., & Kalogirou, S. A. (2008). Artificial Intelligence Techniques for Photovoltaic Applications: A Review. Progress in Energy and Combustion Science, 34, 574-632. http://www.sciencedirect.com/science/article/pii/S0360128508000026
[23]
Monteiro, M. A. (2001). Caracterizacao climatica do estado de santa catarina: Uma abordagem dos principais sistemas atmosfericos que atuam durante o ano. Geosul, 16, 69-78.
[24]
Paletta, Q., & Lasenby, J. (2020). Convolutional Neural Networks Applied to Sky Images for Short-Term Solar Irradiance Forecasting. arXiv:2005.11246 https://arxiv.org/abs/2005.11246
[25]
Pelland, S., Remund, J., Kleissl, J., Oozeki, T., & De Brabandere, K. (2013). Photovoltaic and Solar Forecasting: State of the Art. https://iea-pvps.org/wp-content/uploads/2013/10/Photovoltaic_and_Solar_Forecasting_State_of_the_Art_REPORT_PVPS__T14_01_2013.pdf
[26]
Piccardi, M. (2004). Background Subtraction Techniques: A Review. In 2004 IEEE International Conference on Systems, Man and Cybernetics (Vol. 4, pp. 3099-3104). IEEE. https://doi.org/10.1109/ICSMC.2004.1400815
[27]
Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation. In N. Navab, J. Hornegger, W. Wells, & A. Frangi (Eds.), Medical Image Computing and Computer-Assisted Intervention. Lecture Notes in Computer Science (Vol. 9351, pp. 234-241). Springer. https://doi.org/10.1007/978-3-319-24574-4_28
[28]
Smith, L. N. (2017). Cyclical Learning Rates for Training Neural Networks. In 2017 IEEE Winter Conference on Applications of Computer Vision (WACV) (pp. 464-472). IEEE. https://doi.org/10.1109/WACV.2017.58
[29]
Souza-Echer, M. P., Pereira, E. B., Bins, L., & Andrade, M. A. R. (2006). A Simple Method for the Assessment of the Cloud Cover State in High-Latitude Regions by a Ground-Based Digital Camera. Journal of Atmospheric and Oceanic Technology, 23, 437-447. https://doi.org/10.1175/JTECH1833.1
[30]
Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. https://arxiv.org/abs/1905.11946
[31]
Tarrojam, B., Mueller, F., Eichman, J. D., & Samuelsen, S. (2012). Metrics for Evaluating the Impacts of Intermittent Renewable Generation on Utility Load-Balancing. Energy, 42, 546-562. http://www.sciencedirect.com/science/article/pii/S0360544212001351
[32]
Voyant, C., Notton, G., Kalogirou, S., Nivet, M., Paoli, C., Motte, F., & Fouilloy, A. (2017). Machine Learning Methods for Solar Radiation Forecasting: A Review. Renewable Energy, 105, 569-582. https://www.sciencedirect.com/science/article/pii/S0960148116311648
[33]
Wolf, T., Debut, L., Sanh, V., Chaumond, J., Delangue, C., Moi, A. et al. (2020). Transformers: State-of-the-Art Natural Language Processing. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations (pp. 38-45). Association for Computational Linguistics. https://aclanthology.org/2020.emnlp-demos.6
[34]
Wu, Y., Kirillov, A., Massa, F., Lo, W.-Y., & Girshick, R. (2019). Detectron2. https://github.com/facebookresearch/detectron2
[35]
Ye, L., Cao, Z., Xiao, Y., & Yang, Z. (2019). Supervised Fine-Grained Cloud Detection and Recognition in Whole-Sky Images. IEEE Transactions on Geoscience and Remote Sensing, 57, 7972-7985. https://doi.org/10.1109/TGRS.2019.2917612
[36]
Yuan, Y., Chen, X., & Wang, J. (2020). Object-Contextual Representations for Semantic Segmentation. In A. Vedaldi, H. Bischof, T. Brox, & J. M. Frahm (Eds.), Computer Vision. Lecture Notes in Computer Science (Vol. 12351, pp. 173-190). Springer. https://doi.org/10.1007/978-3-030-58539-6_11
