A novel sky image-based solar irradiance nowcasting model with convolutional block attention mechanism DOI Creative Commons
Shaojian Song, Zijun Yang, Hui Hwang Goh

и другие.

Energy Reports, Год журнала: 2022, Номер 8, С. 125 - 132

Опубликована: Фев. 25, 2022

Global horizontal irradiance (GHI) is a crucial factor impacting photovoltaic (PV) production, and required for accurate real-time power forecasting. And it new effective solution to obtain the GHI by sky images because mainly affected cloud cover motion. Therefore, research proposes unique artificial intelligence approach forecasting ('nowcasting') based on images, which can significantly enhance accuracy cloudy days. First, nowcasting model with convolutional block attention module (CBAM) proposed, Visual Geometry Group (VGG) networks. Then, taking local (LCC) as numerical feature, we coupled feature in image improve performance of model. Finally, verify effectiveness advantages proposed method, when compared state-of-the-art methods, such Sun's model, Jiang's others, method outperforms them demonstrated 11.67% nRMSE, 7.97% nMAE, 27.69% MAPE, 0.91 CORR results ASI-16 dataset.

Язык: Английский

The spatial and temporal mismatch phenomenon in solar space heating applications: status and solutions DOI

Datong Gao,

Bin Zhao, Trevor Hocksun Kwan

и другие.

Applied Energy, Год журнала: 2022, Номер 321, С. 119326 - 119326

Опубликована: Май 31, 2022

Язык: Английский

Процитировано

26
Renewable Energy Maximization for Pelagic Islands Network of Microgrids Through Battery Swapping Using Deep Reinforcement Learning DOI Creative Commons
Muhammad Amin,

Ahmad Suleman,

Muhammad Waseem

и другие.

IEEE Access, Год журнала: 2023, Номер 11, С. 86196 - 86213

Опубликована: Янв. 1, 2023

This paper proposed a reinforcement learning (RL) based energy management system of pelagic islands network microgrids (PINMGs) by ship swapping under the influence environmental impacts. In addition, day-ahead standard scheduling proposing novel method to maximize usage renewable (RE) proposes energy-sharing structure between islands. Energy sharing among plays an important role in electrifying remote islands, which need due unavailability resources meet local demand. The two-stage cooperative multi-agent deep RL has been presented with Q-learning (DQN) approach central and island agents (IA) distributed over numerous overcome this challenge. RL-based approaches efficiently learn optimize their behaviors through several epochs compared other machine or conventional methods in-depth capability. Hence, centralized problem using dueling DQN was solved schedule charge battery from resource-rich (SI) load networks (LIN). case study accuracy different further on because its accurate tracking. Due fluctuating demand charging patterns, for LIN is also stochastic. simulation results, including ship, are validated maximizing RE usefulness algorithm verified state action perturbation verify robustness.

Язык: Английский

Процитировано

14
AI-Driven precision in solar forecasting: Breakthroughs in machine learning and deep learning DOI Creative Commons
Ayesha Nadeem, Muhammad Farhan Hanif,

Muhammad Sabir Naveed

и другие.

AIMS Geosciences, Год журнала: 2024, Номер 10(4), С. 684 - 734

Опубликована: Янв. 1, 2024

<p>The need for accurate solar energy forecasting is paramount as the global push towards renewable intensifies. We aimed to provide a comprehensive analysis of latest advancements in forecasting, focusing on Machine Learning (ML) and Deep (DL) techniques. The novelty this review lies its detailed examination ML DL models, highlighting their ability handle complex nonlinear patterns Solar Irradiance (SI) data. systematically explored evolution from traditional empirical, including machine learning (ML), physical approaches these advanced delved into real-world applications, discussing economic policy implications. Additionally, we covered variety image-based, statistical, ML, DL, foundation, hybrid models. Our revealed that models significantly enhance accuracy, operational efficiency, grid reliability, contributing benefits supporting sustainable policies. By addressing challenges related data quality model interpretability, underscores importance continuous innovation techniques fully realize potential. findings suggest integrating with offers most promising path forward improving forecasting.</p>

Язык: Английский

Процитировано

5
Adaptive co-optimization of artificial neural networks using evolutionary algorithm for global radiation forecasting DOI
Fatih Kılıç, İbrahim Halil Yılmaz,

Özge Kaya

и другие.

Renewable Energy, Год журнала: 2021, Номер 171, С. 176 - 190

Опубликована: Фев. 21, 2021

Язык: Английский

Процитировано

32
A novel sky image-based solar irradiance nowcasting model with convolutional block attention mechanism DOI Creative Commons
Shaojian Song, Zijun Yang, Hui Hwang Goh

и другие.

Energy Reports, Год журнала: 2022, Номер 8, С. 125 - 132

Опубликована: Фев. 25, 2022

Global horizontal irradiance (GHI) is a crucial factor impacting photovoltaic (PV) production, and required for accurate real-time power forecasting. And it new effective solution to obtain the GHI by sky images because mainly affected cloud cover motion. Therefore, research proposes unique artificial intelligence approach forecasting ('nowcasting') based on images, which can significantly enhance accuracy cloudy days. First, nowcasting model with convolutional block attention module (CBAM) proposed, Visual Geometry Group (VGG) networks. Then, taking local (LCC) as numerical feature, we coupled feature in image improve performance of model. Finally, verify effectiveness advantages proposed method, when compared state-of-the-art methods, such Sun's model, Jiang's others, method outperforms them demonstrated 11.67% nRMSE, 7.97% nMAE, 27.69% MAPE, 0.91 CORR results ASI-16 dataset.

Язык: Английский

Процитировано

19