Ultra-short-term Irradiation Prediction Based on Ground-based Cloud Images and Deep Learning DOI
Huiying Fan,

Su Guo

2021 IEEE 5th Conference on Energy Internet and Energy System Integration (EI2), Journal Year: 2023, Volume and Issue: unknown, P. 3540 - 3546

Published: Dec. 15, 2023

Photovoltaic (PV) power generation has been widely used due to its advantages of green and clean, easy installation. However, since output is mainly determined by irradiation, the intermittency, randomness, instability irradiation make PV large-scale grid-connectedness a lousy impact on safety economic operation grid. Therefore, prediction can suppress randomness instability, indirectly improving quality generation. In this paper, firstly, image segmentation processing feature extraction are carried out ground-based cloud images digitize in low dimensions, where new adaptive threshold method based RGB-Grey-OTSU proposed three-valued, separating sun, clouds, sky, comparing it with traditional method. Feature processed images. three values weather, percentage extracted, strong correlation between extracted verified. This paper adopts compares two deep-learning models, LSTM GRU, their performance ultrashort-term different time scales from 5 minutes 1 hour.

Language: Английский

A novel deep learning-based method for theoretical power fitting of photovoltaic generation DOI

Jierui Li,

Xiaoying Ren, Fei Zhang

et al.

Renewable Energy, Journal Year: 2025, Volume and Issue: 250, P. 123271 - 123271

Published: May 5, 2025

Language: Английский

Citations

0
A generative adversarial learning strategy for spatial inspection of compaction quality DOI
Jianhua Li, Xuefei Wang, Jiale Li

et al.

Advanced Engineering Informatics, Journal Year: 2024, Volume and Issue: 62, P. 102791 - 102791

Published: Sept. 2, 2024

Language: Английский

Citations

3
Generative Adversarial Networks for Synthetic Meteorological Data Generation DOI

Diogo Viana,

Rita Teixeira, Tiago Soares

et al.

Lecture notes in computer science, Journal Year: 2024, Volume and Issue: unknown, P. 197 - 206

Published: Nov. 15, 2024

Language: Английский

Citations

0
Daily electric vehicle charging dataset for training reinforcement learning algorithms DOI Creative Commons
Nastaran Gholizadeh, Petr Musı́lek

Data in Brief, Journal Year: 2024, Volume and Issue: 55, P. 110587 - 110587

Published: June 3, 2024

Reinforcement learning algorithms are increasingly utilized across diverse domains within power systems. One notable challenge in training and deploying these is the acquisition of large, realistic datasets. It imperative that trained on extensive, datasets over numerous iterations to ensure optimal performance real-world scenarios. In pursuit this goal, we curated a comprehensive dataset capturing electric vehicle (EV) charging details span 29,600 days designated parking facility. This encompasses necessary information such as connection times, durations, energy consumption individual EVs. The methodology involved employing conditional tabular generative adversarial networks (CTGAN) craft pool synthetic from smaller initial collected an EV facility located Caltech campus. Subsequently, multiple post-processing techniques were implemented extract data pool, ensuring compliance with station's capacity constraint while maintaining daily demand profile derived historical data. Using kernel density estimation (KDE), distributional characteristics data, especially concerning timing connections, faithfully replicated. developed specifically useful offline reinforcement algorithms.

Language: Английский

Citations

0
Modeling Dissolved Oxygen Under Data Scarcity Situation Using Time-Series Generative Adversarial Network Combined with Long Short-Term Memory Network DOI
Gang Li, Cheng Chen, Siyang Yao

et al.

Published: Jan. 1, 2024

Language: Английский

Citations

0
A method for predicting methane production from anaerobic digestion of kitchen waste under small sample conditions DOI
Shipin Yang,

Yuqiao Cai,

Tingting Zhao

et al.

Environmental Science and Pollution Research, Journal Year: 2024, Volume and Issue: 31(37), P. 49615 - 49625

Published: July 30, 2024

Language: Английский

Citations

0
Ultra-short-term Irradiation Prediction Based on Ground-based Cloud Images and Deep Learning DOI
Huiying Fan,

Su Guo

2021 IEEE 5th Conference on Energy Internet and Energy System Integration (EI2), Journal Year: 2023, Volume and Issue: unknown, P. 3540 - 3546

Published: Dec. 15, 2023

Photovoltaic (PV) power generation has been widely used due to its advantages of green and clean, easy installation. However, since output is mainly determined by irradiation, the intermittency, randomness, instability irradiation make PV large-scale grid-connectedness a lousy impact on safety economic operation grid. Therefore, prediction can suppress randomness instability, indirectly improving quality generation. In this paper, firstly, image segmentation processing feature extraction are carried out ground-based cloud images digitize in low dimensions, where new adaptive threshold method based RGB-Grey-OTSU proposed three-valued, separating sun, clouds, sky, comparing it with traditional method. Feature processed images. three values weather, percentage extracted, strong correlation between extracted verified. This paper adopts compares two deep-learning models, LSTM GRU, their performance ultrashort-term different time scales from 5 minutes 1 hour.

Language: Английский

Citations

0