A Hybrid Prediction Model of Photovoltaic Power System Based on AP, Issa-Based Vmd, Clkan and Error Correction

Опубликована: Янв. 1, 2025

The accurate prediction of photovoltaic (PV) power output is crucial for optimal energy management. However, PV generation systems are influenced by various meteorological factors, resulting in the fluctuation and intermittency issues their power. To enhance accuracy power, this study proposes a novel hybrid model, Convolutional Neural Network-Long Short-Term Memory-Kolmogorov-Arnold networks (CLKAN), integrated with Improved Sparrow Search Algorithm (ISSA), Variational Mode Decomposition (VMD), Error Correction (EC). key contributions research as follows: (i) Iterative Chaotic Map Infinite Collapses (ICMIC) method Quantum Rotation Gate (QRG) applied first time to SSA optimization algorithm. This approach improves issue redundant information dispersion sequences due suboptimal VMD hyperparameters. (ii) By leveraging CNN-LSTM input sequence feature extraction, KAN network B-spline basis function innovatively utilized connect CNN-LSTM-KAN model. Experimental analysis conducted on four typical days across different seasons at station Yunnan Province, China, shows that proposed model achieves higher reduced training time. For instance summer, achieved best performance MAE, SMAPE, RMSE, R2 values 0.22 MW, 0.88%, 0.28 99.91%, respectively, demonstrating its superior performance. Furthermore, application CLKAN based AP ISVMD, stations Gansu, highlights model's robustness spatial temporal scales.

Язык: Английский

---

Assessing the Potential Impact of Aerosol Scenarios for Rooftop PV Regional Deployment

Renewable Energy, Год журнала: 2025, Номер unknown, С. 122869 - 122869

Опубликована: Март 1, 2025

Язык: Английский

---

Cost-oriented capacity assessment for renewable power systems with automatic reserve neural networks and information gap decision theory

Energy Reports, Год журнала: 2025, Номер 13, С. 3430 - 3441

Опубликована: Март 14, 2025

Язык: Английский

---

Review on PV uncertainty model

Elsevier eBooks, Год журнала: 2025, Номер unknown, С. 1 - 34

Опубликована: Янв. 1, 2025

Язык: Английский

---

An adaptive ensemble framework using multi-source data for day-ahead photovoltaic power forecasting

Journal of Renewable and Sustainable Energy, Год журнала: 2024, Номер 16(1)

Опубликована: Янв. 1, 2024

Day-ahead photovoltaic (PV) power forecasting plays a crucial role in market trading and grid dispatching. It has been empirically demonstrated various fields that combining forecasts yields better results than using individual models. In this work, novel adaptive ensemble framework is proposed based on multi-source data. First, incorporating prior information from physical models, three types of high-performance component models are constructed different Second, multi-label classification method utilized to select performing allowing for switching between model combinations depending the weather conditions. Finally, dynamic used update weights its cumulative errors observed recent past. The was evaluated four-year PV dataset 2019 2022. skill (FS) test year (2022) reaches 50.61%. show FS improved by 4.75% compared optimal model. Compared with other state-of-the-art methods, our achieved best performance improving at least 3.94%. study can be widely applied energy fields, such as wind/load forecasting.

Язык: Английский

---

A correction framework for day-ahead NWP solar irradiance forecast based on sparsely activated multivariate-shapelets information aggregation

Renewable Energy, Год журнала: 2025, Номер unknown, С. 122638 - 122638

Опубликована: Фев. 1, 2025

Язык: Английский

---

Accelerating convolutional neural networks on FPGA platforms: a high-performance design methodology using OpenCL

Journal of Real-Time Image Processing, Год журнала: 2025, Номер 22(2)

Опубликована: Фев. 26, 2025

Язык: Английский

---

Photovoltaic power forecasting based on VMD-SSA-Transformer: Multidimensional analysis of dataset length, weather mutation and forecast accuracy

Energy, Год журнала: 2025, Номер unknown, С. 135971 - 135971

Опубликована: Апрель 1, 2025

Язык: Английский

---

AI-Based Disaster Prediction and Early Warning Systems

Advances in computational intelligence and robotics book series, Год журнала: 2025, Номер unknown, С. 55 - 74

Опубликована: Фев. 28, 2025

Disaster management systems. This introduction explores the role of disaster technologies, focusing on their evolution, current applications, and future potential, particularly in context AI-based solutions.Historically, relied heavily manual processes rudimentary tools for assessment response. often resulted delayed reactions inadequate resource allocation during emergencies. Geographic Information Systems (GIS), remote sensing, telecommunications have all played crucial roles improving situational awareness enabling faster decision-making.ICT facilitates real-time data sharing among stakeholders, including government agencies, NGOs, public. connectivity enhances coordination ensures that critical information reaches affected populations promptly. For instance, mobile technology has empowered communities to report incidents receive alerts, thereby early warning systems.Moreover, analytics emerged as a key component management.

Язык: Английский

---

A deep learning framework for photovoltaic power forecasting in multiple interconnected countries

Sustainable Energy Technologies and Assessments, Год журнала: 2025, Номер 77, С. 104330 - 104330

Опубликована: Апрель 26, 2025

Язык: Английский

---