Related references
Note: Only part of the references are listed.
Review
Economics
Fan Lin et al.
Summary: As the use of solar energy in power systems increases, accurate intra-hour solar forecasting (IHSF) becomes crucial for secure and economical power system operation. The main challenge in achieving high accuracy IHSF is the rapid changes in clouds. Ground-based sky image (GSI) provides intuitive information on cloud changes and has been widely used in IHSF studies. This paper presents a comprehensive review of the state-of-the-art in ground-based sky image-based intra-hour solar forecasting (GSI-IHSF) and proposes a generic framework consisting of four modules: sky image acquisition, preprocessing, cloud forecasting, and solar forecasting. It also discusses research trends, recent advances, potential improvements in forecast accuracy, and identifies future requirements and challenges.
INTERNATIONAL JOURNAL OF FORECASTING
(2023)
Article
Energy & Fuels
Luis Eduardo Ordonez Palacios et al.
Summary: Understanding solar energy is crucial for modern societies, and this study built predictive models of solar energy at different altitudes using satellite imagery and solar radiation data. The Random Forest algorithm showed the best performance in the model below 800 m.a.s.l.
Article
Green & Sustainable Science & Technology
Xiaoyang Chen et al.
Summary: This paper introduces the applicability of solar nowcasting for the first time, exploring its impact on the PV power ramp-rate control algorithm. The results suggest that the forecast horizon has the most significant impact on applicability, while the forecast model updating rate has the least. Moreover, negatively-biased deterministic nowcasts and wider probabilistic nowcasts are found to be more applicable to PV power ramp-rate control.
Article
Energy & Fuels
Shaojian Song et al.
Summary: This study proposes a real-time global horizontal irradiance (GHI) forecasting method based on sky images, using artificial intelligence and convolutional neural network models. By combining the local cloud cover as a numerical feature with the cloud feature in the sky image, the forecasting accuracy of GHI on cloudy days can be significantly enhanced. Experimental results demonstrate that the proposed method outperforms other state-of-the-art methods in GHI forecasting.
Article
Energy & Fuels
Quentin Paletta et al.
Summary: The article introduces a spatio-temporal neural network architecture called ECLIPSE for predicting the temporal variability of solar irradiance. The model can not only predict future irradiance levels and uncertainties, but also generate segmented images to provide richer information.
Article
Energy & Fuels
Hui-Min Zuo et al.
Summary: In this study, a short-term solar irradiance prediction model based on a deep learning network was proposed. By analyzing the relationship between cloud coverage and future solar irradiance, combined with meteorological parameters and historical data, the model achieved better performance compared to other models.
Review
Engineering, Electrical & Electronic
Cong Feng et al.
Summary: The exponential growth of solar power in the past decade highlights its potential, but also brings challenges. The rise of smart grids enables the use of AI techniques to address solar integration issues. While different AI subfields have made breakthroughs in solar energy, research in this area is still at an early stage and lags behind mainstream AI development.
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS
(2021)
Article
Automation & Control Systems
Haoran Wen et al.
Summary: This article introduces a novel multistep forecasting approach for PV power ramp-rate control utilizing deep ConvNets, which can make multiple predictions with high temporal resolution. By incorporating stacked sky images, the forecasting performance is further improved. Experimental results show that the MSF-based PRRC can detect more ramp-rates violations with a higher control rate compared to conventional methods.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Chemistry, Analytical
Raoof Naushad et al.
Summary: The study implemented transfer learning to fine-tune VGG16 and WRNs networks for LULC classification using the EuroSAT dataset, achieving high accuracy while addressing the limited-data problem. Techniques such as early stopping, gradient clipping, adaptive learning rates, and data augmentation were used to optimize performance and computational time. The proposed approach based on WRNs outperformed previous methods in terms of computational efficiency and accuracy, achieving 99.17% accuracy.
Article
Energy & Fuels
Quentin Paletta et al.
Summary: This study compares the performance of four deep learning architectures for solar irradiance forecasting, finding that encoding spatiotemporal aspects of the sky image sequence greatly improves predictions. However, deep learning models tend to behave like a 'very smart persistence model', missing fundamental events causing large irradiance changes such as cloud cover. The hope is for this work to shift the approach to irradiance forecasting from reactive to anticipatory.
Article
Chemistry, Multidisciplinary
Rial A. Rajagukguk et al.
Summary: This study attempted to forecast cloud cover using deep learning models, which were then used to predict solar irradiance. The data collected from sky images were processed to develop a deep learning model for forecasting cloud cover 10 minutes ahead. The forecasted cloud cover data were applied to solar radiation models to predict global horizontal irradiance.
APPLIED SCIENCES-BASEL
(2021)
Article
Meteorology & Atmospheric Sciences
Bu-Yo Kim et al.
Summary: This study used image data features and machine learning methods to calculate 24-hour continuous cloud cover from ground-based image data. The support vector regression model achieved the best results, showing potential for replacing human-eye observations.
ATMOSPHERIC MEASUREMENT TECHNIQUES
(2021)
Article
Multidisciplinary Sciences
Alessandro Niccolai et al.
Review
Computer Science, Artificial Intelligence
Mario P. Vestias et al.
Article
Computer Science, Information Systems
Huaiguang Jiang et al.
Article
Environmental Sciences
Quang-Khai Tran et al.
Article
Green & Sustainable Science & Technology
Hugo T. C. Pedro et al.
JOURNAL OF RENEWABLE AND SUSTAINABLE ENERGY
(2019)
Article
Computer Science, Information Systems
Henrik Zsiboracs et al.
Article
Energy & Fuels
A. Ayet et al.
Review
Energy & Fuels
Steven D. Miller et al.
Article
Green & Sustainable Science & Technology
Walter Richardson et al.
Article
Green & Sustainable Science & Technology
Yinghao Chu et al.
Article
Engineering, Ocean
Qingyong Li et al.
JOURNAL OF ATMOSPHERIC AND OCEANIC TECHNOLOGY
(2011)
