A Preprocessing and Feature Extraction Method of Ground-based Cloud Images for Photovoltaic Power Prediction

Lu, Zhiying; Wang, Kaixuan; Li, Xin

doi:10.23919/ccc52363.2021.9549239

Cited by 2 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, the ground-based cloud map is transformed from the RGB spatial color model to the HSV spatial color model, and the cloud thickness is accurately quantified using the V component [7] . The equations for each component of the HSV model are as follows.…”

Section: Extraction Of Cloud Thicknessmentioning

confidence: 99%

“…The all-sky imager is the mainstream equipment currently used for minute-scale PV prediction, and scholars at home and abroad use it to photograph cloud masses in the sky to obtain intuitive cloud features, based on which the ultra-short-term PV power is predicted. Literature [7] used ground-based cloud map to extract image features such as light intensity, transmittance, and cloud factor for PV power prediction using gradient boosting decision tree (GBDT) model, and literature [8] used digital image processing techniques to extract radiation-related image features for PV power prediction by radial basis function neural network, but none of the above literature considered the motion of cloud masses; literature [9] used cloud mass extraction algorithm and tracking learning algorithm to achieve the prediction of future cloud motion, but treating cloud motion as translation will reduce the accuracy of cloud trajectory prediction; literature [10] proposed a cloud trajectory tracking method based on HSV-SURF features when extracting ground-based cloud map features, but did not consider static features such as light intensity and cloud shape.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

PV power prediction method based on ground-based cloud map and hybrid neural network

Ruiqi,

Gang,

Jianping

et al. 2023

Fourth International Conference on Artificial Intelligence and Electromechanical Automation (AIEA 2023)

View full text Add to dashboard Cite

Under cloudy weather, PV power will fluctuate dramatic due to the sporty cloud shading. However, most existing PV power prediction models do not utilize the cloud map sufficiently. In this paper, we propose a novel method to improve the prediction accuracy through combining cloud map features and hybrid neural network. Firstly, we extract the static and dynamic features of the ground-based cloud map using image processing techniques. Secondly, we establish an PV power prediction model based on the ensemble empirical mode decomposition-bi-long short-term memory (EEMD-BiLSTM) with numerical weather information and historical power data. Finally, in order to further improve the power prediction accuracy of cloudy weather, we build an error correction model using the cloud map features, based on the light gradient boosting machine (LightGBM) decision tree. Experimental results on the data from a PV plant in western China shows that our method can effectively improve the accuracy of PV power prediction under cloudy weather.

show abstract

Section: Extraction Of Cloud Thicknessmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

PV power prediction method based on ground-based cloud map and hybrid neural network

Ruiqi,

Gang,

Jianping

et al. 2023

Fourth International Conference on Artificial Intelligence and Electromechanical Automation (AIEA 2023)

View full text Add to dashboard Cite

show abstract

“…The ground-based cloud prediction method analyses cloud information collected by ground-based all-sky imagers in order to assess the effect of clouds on solar radiation and to predict PV power (Sun et al, 2014). This method is conducive to improving the accuracy of PV power prediction, especially in short-and ultrashort-term prediction, and has obvious advantages (Jaouhari, Zaz and Masmoudi, 2015;Lu, Wang and Li, 2021). However, the ground-based cloud mapping prediction method has limitations in temporal information utilization and is weak for migration between PV systems (Wei et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Photovoltaic power prediction based on sky images and tokens-to-token vision transformer

Dai,

Huo,

et al. 2023

Int. J. Renew. Energy Dev.

View full text Add to dashboard Cite

Photovoltaic (PV) power generation has high uncertainties due to the randomness and imbalance nature of solar energy and meteorological parameters. Hence, accurate PV power forecasts are essential in the operation of PV power plants (PVPP) for short-term dispatches and power generation schedules. In this paper, a new deep neural network structure based on vision transformer is proposed to combine sky images and Tokens-To-Token（T2T） for photovoltaic power prediction. The method uses an incremental tokenization module to aggregate neighboring image patches into tokens, which capture the local structural information of the clouds. Then, an efficient T2T-ViT backbone network is used to extract the global attentional relationships of the tokens for power prediction. In order to evaluate the performance of the proposed model, the method was compared with several deep learning architectures such as ResNet and GoogleNet on a dataset collected by the National Renewable Energy Laboratory in Colorado, USA. The results of power prediction were analysed using training loss, prediction error, and linear regression, and they show that the proposed method achieves higher prediction accuracy and lower error compared to the existing methods, especially in short- and ultra-short-term prediction. The paper demonstrates the potential of applying Transformer models to computer vision tasks for renewable energy forecasting. The results show that the proposed method achieves higher prediction accuracy and lower error than several deep learning architectures, such as ResNet and GoogleNet, especially in short- and ultra-short-term prediction.

show abstract

A Preprocessing and Feature Extraction Method of Ground-based Cloud Images for Photovoltaic Power Prediction

Cited by 2 publications

References 11 publications

PV power prediction method based on ground-based cloud map and hybrid neural network

PV power prediction method based on ground-based cloud map and hybrid neural network

Photovoltaic power prediction based on sky images and tokens-to-token vision transformer

Contact Info

Product

Resources

About