Spatio-Temporal Photovoltaic Power Prediction with Fourier Graph Neural Network DOI Open Access
Jing Shi,

Xianpeng Xi,

Dongdong Su

et al.

Electronics, Journal Year: 2024, Volume and Issue: 13(24), P. 4988 - 4988

Published: Dec. 18, 2024

The strong development of distributed energy sources has become one the most important measures for low-carbon worldwide. With a significant quantity photovoltaic (PV) power generation being integrated to grid, accurate and efficient prediction PV is an essential guarantee security stability electricity grid. Due shortage data from stations influence weather, it difficult obtain satisfactory performance prediction. In this regard, we present forecasting model based on Fourier graph neural network (FourierGNN). Firstly, hypervariable constructed by considering weather neighbouring plants as nodes, respectively. hypervariance then transformed in space capture spatio-temporal dependence among nodes via discrete transform. multilayer operator (FGO) can be further exploited information. Experiments carried out at six show that presented approach enables optimal obtained adequately exploiting

Language: Английский

Short-term Interval Prediction Strategy of Photovoltaic Power Based on Meteorological Reconstruction with Spatiotemporal Correlation and Multi-factor Interval Constraints DOI

Mao Yang,

Yue Jiang, Wei Zhang

et al.

Renewable Energy, Journal Year: 2024, Volume and Issue: 237, P. 121834 - 121834

Published: Nov. 6, 2024

Language: Английский

Citations

23
Short-Term Photovoltaic Power Forecasting Based on the VMD-IDBO-DHKELM Model DOI Creative Commons
Shengli Wang, Xiaolong Guo, Tian-Le Sun

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(2), P. 403 - 403

Published: Jan. 17, 2025

A short-term photovoltaic power forecasting method is proposed, integrating variational mode decomposition (VMD), an improved dung beetle algorithm (IDBO), and a deep hybrid kernel extreme learning machine (DHKELM). First, the weather factors less relevant to (PV) generation are filtered using Spearman correlation coefficient. Historical data then clustered into three categories—sunny, cloudy, rainy days—using K-means algorithm. Next, original PV decomposed through VMD. DHKELM-based combined prediction model developed for each component of decomposition, tailored different types. The model’s hyperparameters optimized IDBO. final forecast determined by combining outcomes individual component. Validation performed actual from plant in Australia station Kashgar, China demonstrates. Numerical evaluation results show that proposed improves Mean Absolute Error (MAE) 3.84% Root-Mean-Squared (RMSE) 3.38%, confirming its accuracy.

Language: Английский

Citations

0
Adaptive Robust Optimal Scheduling of Combined Heat and Power Microgrids Based on Photovoltaic Mechanism/Data Fusion-Driven Power Prediction DOI Creative Commons

Yueyang Xu,

Yibo Wang, Chuang Liu

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(3), P. 732 - 732

Published: Feb. 5, 2025

In order to effectively deal with the adverse effects of randomness photovoltaic output on operation combined heat and power (CHP) microgrids, this paper proposes an adaptive robust optimal scheduling strategy for CHP microgrids based mechanism/data fusion-driven prediction. Firstly, mechanism clear sky radiation model is used calculate limit random output, latter reorganized in different periods by using idea similar days. Then, data-driven prediction results are superimposed established, framework provided. Secondly, boundary information uncertain factors deeply explored, uncertainty set considering confidence interval predictive error statistical constructed. On basis, a optimization lowest operating cost proposed, solved column constraint generation algorithm. Finally, rationality effectiveness proposed verified through simulation examples analytical calculations.

Language: Английский

Citations

0
Fault Diagnosis of Photovoltaic Array Based on Improved Honey Badger Optimization Algorithm DOI Creative Commons

Ziwei Guo,

Yuanyuan Chang,

Yanhong Fang

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(4), P. 841 - 841

Published: Feb. 11, 2025

A photovoltaic array fault diagnosis method based on an improved honey badger optimization algorithm is proposed to improve the accuracy of diagnosis. Firstly, analyze current and power output characteristic curves under different states, construct a preliminary set 10 dimensional feature vectors. Secondly, vector ranked in importance using random forest algorithm, then input into support machines, long short-term memory, bidirectional memory neural networks obtain optimal combination base model number features. Then, was by combining Tent chaotic mapping column measurement, control factors, pinhole imaging strategy, compared with other algorithms demonstrate its effectiveness ability, stability, convergence speed. Finally, features, problem hyperparameter setting effectively solved. The experimental results show that 97.1014%, which superior models verifies method.

Language: Английский

Citations

0
Improved bidirectional long short-term memory network-based short-term forecasting of photovoltaic power for different seasonal types and weather factors DOI

Ruixian Wang,

Rui Ma,

Linjun Zeng

et al.

Computers & Electrical Engineering, Journal Year: 2025, Volume and Issue: 123, P. 110219 - 110219

Published: March 1, 2025

Language: Английский

Citations

0
A short-term power prediction method based on the transformation of multi-source spatiotemporal feature for photovoltaic cluster DOI

Chaohong Zhou,

Fan Zhang,

Shenhui Gu

et al.

International Journal of Green Energy, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 17

Published: March 3, 2025

Language: Английский

Citations

0
Enhancing short-term algal bloom forecasting through an anti-mimicking hybrid deep learning method DOI
Ya-Qin Zhang,

Yichong Wang,

Huihuang Chen

et al.

Journal of Environmental Management, Journal Year: 2025, Volume and Issue: 379, P. 124832 - 124832

Published: March 10, 2025

Language: Английский

Citations

0
Research on Photovoltaic Power Prediction Method Based on Dynamic Similar Selection and Bidirectional Gated Recurrent Unit DOI Open Access
Qinghong Wang, Longhao Li

Advanced Theory and Simulations, Journal Year: 2025, Volume and Issue: unknown

Published: March 8, 2025

Abstract Photovoltaic (PV) power generation is vital for sustainable energy development, yet its inherent randomness and volatility challenge grid stability. Accurate short‐term PV prediction essential reliable operation. This paper proposes an integrated method combining dynamic similar selection (DSS), variational mode decomposition (VMD), bidirectional gated recurrent unit (BiGRU), improved sparrow search algorithm (ISSA). First, DSS selects training data based on local meteorological similarity, reducing interference. VMD then decomposes into smooth components, mitigating volatility. The Pearson correlation coefficient used to filter highly relevant variables, enhancing input quality. BiGRU captures temporal evolution patterns, with ISSA optimizing key parameters robust forecasting. Validated historical Australian under diverse weather conditions, the proposed effectively reduces volatility, significantly improving accuracy reliability. These advancements support stable supply efficient

Language: Английский

Citations

0
A deep learning model based on multi-attention mechanism and gated recurrent unit network for photovoltaic power forecasting DOI
Kuo Yang,

Yanjie Cai,

Junfang Cheng

et al.

Computers & Electrical Engineering, Journal Year: 2025, Volume and Issue: 123, P. 110250 - 110250

Published: March 24, 2025

Language: Английский

Citations

0
Fault Diagnosis Method of Rolling Bearing Based on 1D Multi-Channel Improved Convolutional Neural Network in Noisy Environment DOI Creative Commons

Huijuan Guo,

Dongzhi Ping,

Lijun Wang

et al.

Sensors, Journal Year: 2025, Volume and Issue: 25(7), P. 2286 - 2286

Published: April 4, 2025

The vibration signal of mechanical equipment in operating environments is the key to describing fault characteristics, but due thez influence density and environmental interference, accuracy diagnosis often affected by noise. In this paper, a method based on 1D Multi-Channel Improved Convolutional Neural Network (1DMCICNN) proposed. By introducing BiLSTM, an attention mechanism local sparse structure two-channel Network, feature information noisy timing fully extracted at different scales while reducing computational parameters. model verified through experiments under signal-to-noise ratios loads. results show that 1DMCICNN 98.67%, 99.71%, 99.04%, 99.71% load speed datasets. Meanwhile, compared with unoptimized training parameters are reduced 55.58%.

Language: Английский

Citations

0