CEEMDAN-RIME–Bidirectional Long Short-Term Memory Short-Term Wind Speed Prediction for Wind Farms Incorporating Multi-Head Self-Attention Mechanism DOI Creative Commons
Wenlu Yang, Zhanqiang Zhang, Keqilao Meng

и другие.

Applied Sciences, Год журнала: 2024, Номер 14(18), С. 8337 - 8337

Опубликована: Сен. 16, 2024

Accurate wind speed prediction is extremely critical to the stable operation of power systems. To enhance accuracy, we propose a new approach that integrates bidirectional long short-term memory (BiLSTM) with fully adaptive noise ensemble empirical modal decomposition (CEEMDAN), RIME optimization algorithm (RIME), and multi-head self-attention mechanism (MHSA). First, historical data farms are decomposed via CEEMDAN extract change patterns features on different time scales, subsequences obtained. Then, parameters BiLSTM model optimized using frost ice algorithm, each subsequence input into neural network containing MHSA for prediction. Finally, predicted values component weighted reconstructed obtain series. According experimental results, method can predict speeds more accurately. We verified effectiveness by comparing it models.

Язык: Английский

Artificial hummingbird algorithm: Theory, variants, analysis, applications, and performance evaluation DOI
Buddhadev Sasmal, Arunita Das, Krishna Gopal Dhal

и другие.

Computer Science Review, Год журнала: 2025, Номер 56, С. 100727 - 100727

Опубликована: Янв. 18, 2025

Язык: Английский

Процитировано

2
A multi-factor clustering integration paradigm for wind speed point-interval prediction based on feature selection and optimized inverted transformer DOI
Jujie Wang, Weiyi Jiang, Shuqin Shu

и другие.

Energy, Год журнала: 2025, Номер unknown, С. 135210 - 135210

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

1
Research on large-scaled wire-bonding machine scheduling in SAT: EAHA with knowledge learning and progressive fusion decomposition DOI
Hong Wang, Da Chen, Lihui Wu

и другие.

Enterprise Information Systems, Год журнала: 2025, Номер unknown

Опубликована: Янв. 30, 2025

Язык: Английский

Процитировано

0
AI-Enabled Predictive Analytics in Smart Grids: The Case of Sweden DOI Creative Commons
Theodore Kindong, Björn Johansson, Victoria Paulsson

и другие.

Complex Systems Informatics and Modeling Quarterly, Год журнала: 2025, Номер 42, С. 43 - 62

Опубликована: Апрель 30, 2025

Smart grids (SGs) revolutionize existing power by using a wide range of developing disruptive technologies to generate clean, efficient, and predictable energy. Our study uses an action research method focuses solely on the first two stages process, diagnosis planning, evaluate ways adopt artificial intelligence (AI) applications in SGs for predictive analytics practice. The stage entails conducting systematic literature review AI SGs, highlighting four areas potential analytics: outage prediction, demand response, control coordination, AI-enabled security optimize decision-making, diagnose faults, improve grid stability security. planning step included document analysis devise methods enable practical implementation smart analytics. Finally, we address implementing transparent analytics, followed conclusion future direction. study’s key is that more needed complete taking (implementing solution), evaluation (assessing results), learning (reflecting lessons learned) phases cycle.

Язык: Английский

Процитировано

0
Improving Offshore Wind Speed Forecasting with a CRGWAA-Enhanced Adaptive Neuro-Fuzzy Inference System DOI Creative Commons

Yingjie Liu,

Fahui Miao

Journal of Marine Science and Engineering, Год журнала: 2025, Номер 13(5), С. 908 - 908

Опубликована: Май 3, 2025

Accurate forecasting of offshore wind speed is crucial for the efficient operation and planning energy systems. However, inherently non-stationary highly volatile nature speed, coupled with sensitivity neural network-based models to parameter settings, poses significant challenges. To address these issues, this paper proposes an Adaptive Neuro-Fuzzy Inference System (ANFIS) optimized by CRGWAA. The proposed CRGWAA integrates Chebyshev mapping initialization, elite-guided reflection refinement operator, a generalized quadratic interpolation strategy enhance population diversity, adaptive exploration, local exploitation capabilities. performance comprehensively evaluated on CEC2022 benchmark function suite, where it demonstrates superior optimization accuracy, convergence robustness compared six state-of-the-art algorithms. Furthermore, ANFIS-CRGWAA model applied short-term using real-world data from region Fujian, China, at 10 m 100 above sea level. Experimental results show that consistently outperforms conventional hybrid baselines, achieving lower MAE, RMSE, MAPE, as well higher R2, across both altitudes. Specifically, original ANFIS-WAA model, RMSE reduced approximately 45% 24% m. These findings confirm effectiveness, stability, generalization ability complex, prediction tasks.

Язык: Английский

Процитировано

0
Using stacking ensemble learning to predict multi-step wind speed based on wavelet transformation, two-steps feature selection method, and neural networks DOI
Faezeh Amirteimoury, Gholamreza Memarzadeh, Farshid Keynia

и другие.

Measurement, Год журнала: 2024, Номер 244, С. 116500 - 116500

Опубликована: Дек. 12, 2024

Язык: Английский

Процитировано

2
Algorithm Initialization: Categories and Assessment DOI
Abdul Halim, Swagatam Das, Idris Ismail

и другие.

Emergence, complexity and computation, Год журнала: 2024, Номер unknown, С. 1 - 100

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

1
Using Stacking Ensemble Learning to Predict Multi-Step Wind Speed Based on Wavelet Transformation, a Two-Steps Feature Selection Method, and Neural Networks DOI
Faezeh Amirteimoury, Gholamreza Memarzadeh, Farshid Keynia

и другие.

Опубликована: Янв. 1, 2024

Wind energy is gaining attention in power sector. However, the instability of wind speed (WS) negatively affects incorporation into grid. Reducing issues requires precise WS forecasting. The current paper introduces an ensemble approach for multi-steps forecasting including discrete wavelet transform (DWT) to denoise signal and mutual information-interaction gain (MI-IG) identify most relevant input features. Moreover, multi-layer perceptron (MLP), bidirectional long short-term memory (BiLSTM), gated recurrent unit (GRU), convolutional neural network (CNN) are employed build individual modules. Finally, stacking learning combines outputs from these For first dataset, proposed model achieved MSE between 0.05 0.59, MAE 0.19 0.6, MAPE 5.97% 21.94% R2 0.985 0.852 one-hour ahead five-hours ahead.

Язык: Английский

Процитировано

0
CEEMDAN-RIME–Bidirectional Long Short-Term Memory Short-Term Wind Speed Prediction for Wind Farms Incorporating Multi-Head Self-Attention Mechanism DOI Creative Commons
Wenlu Yang, Zhanqiang Zhang, Keqilao Meng

и другие.

Applied Sciences, Год журнала: 2024, Номер 14(18), С. 8337 - 8337

Опубликована: Сен. 16, 2024

Accurate wind speed prediction is extremely critical to the stable operation of power systems. To enhance accuracy, we propose a new approach that integrates bidirectional long short-term memory (BiLSTM) with fully adaptive noise ensemble empirical modal decomposition (CEEMDAN), RIME optimization algorithm (RIME), and multi-head self-attention mechanism (MHSA). First, historical data farms are decomposed via CEEMDAN extract change patterns features on different time scales, subsequences obtained. Then, parameters BiLSTM model optimized using frost ice algorithm, each subsequence input into neural network containing MHSA for prediction. Finally, predicted values component weighted reconstructed obtain series. According experimental results, method can predict speeds more accurately. We verified effectiveness by comparing it models.

Язык: Английский

Процитировано

0