Optimized Seq2Seq model based on multiple methods for short-term power load forecasting

Applied Soft Computing, Journal Year: 2023, Volume and Issue: 142, P. 110335 - 110335

Published: April 25, 2023

Language: Английский

---

A Bayesian Optimization-Based LSTM Model for Wind Power Forecasting in the Adama District, Ethiopia

Energies, Journal Year: 2023, Volume and Issue: 16(5), P. 2317 - 2317

Published: Feb. 28, 2023

Renewable energies, such as solar and wind power, have become promising sources of energy to address the increase in greenhouse gases caused by use fossil fuels resolve current crisis. Integrating into a large-scale electric grid presents significant challenge due high intermittency nonlinear behavior power. Accurate power forecasting is essential for safe efficient integration system. Many prediction models been developed predict uncertain time series but most neglect Bayesian optimization optimize hyperparameters while training deep learning algorithms. The efficiency search strategies decreases number increases, computation complexity becomes an issue. This paper robust optimized long-short term memory network generation day ahead context Ethiopia’s renewable sector. proposal uses find best hyperparameter combination reasonable time. results indicate that tuning using this metaheuristic prior building significantly improves predictive performances models. proposed were evaluated MAE, RMSE, MAPE metrics, outperformed both baseline gated recurrent unit architecture.

Language: Английский

---

Forecasting carbon price using signal processing technology and extreme gradient boosting optimized by the whale optimization algorithm

Energy Science & Engineering, Journal Year: 2024, Volume and Issue: 12(3), P. 810 - 834

Published: Jan. 2, 2024

Abstract Predicting carbon prices is crucial for the growth of China's trading industry. This paper proposes a residual correction model that considers multiple influencing factors. First, best historical data and main external factors input by are determined using partial autocorrelation function Spearman correlation analysis, price forecasting index system constructed. Second, whale optimization algorithm (WOA) utilized to determine optimal parameters extreme gradient boosting (XGBoost), WOA‐XGBoost built perform preliminary forecasts obtain series. Finally, series undergoes decomposition into components utilizing complete ensemble empirical mode subsequent aggregation outcomes. Experiments conducted predict two markets in Hubei Guangzhou, feature importance analysis performed. The results indicate proposed hybrid consistently outperforms comparative models terms prediction accuracy. Furthermore, it revealed European Union key market prices.

Language: Английский

---

Application of ensemble empirical mode decomposition with support vector regression and wavelet neural network in electric load forecasting

Energy Sources Part B Economics Planning and Policy, Journal Year: 2025, Volume and Issue: 20(1)

Published: Feb. 22, 2025

Language: Английский

---

Building cooling load forecasting of IES considering spatiotemporal coupling based on hybrid deep learning model

Applied Energy, Journal Year: 2023, Volume and Issue: 349, P. 121547 - 121547

Published: July 28, 2023

Language: Английский

---

A local thermal sensation model suitable for thermal comfort evaluation of sensitive body segments

Sustainable Cities and Society, Journal Year: 2023, Volume and Issue: 97, P. 104751 - 104751

Published: June 29, 2023

Language: Английский

---

A control method combining load prediction and operation optimization for phase change thermal energy storage system

Sustainable Cities and Society, Journal Year: 2023, Volume and Issue: 96, P. 104674 - 104674

Published: May 28, 2023

Language: Английский

---

A multi-stage ensemble model for power load forecasting based on decomposition, error factors, and multi-objective optimization algorithm

International Journal of Electrical Power & Energy Systems, Journal Year: 2023, Volume and Issue: 155, P. 109620 - 109620

Published: Nov. 5, 2023

Short-term power load forecasting plays an important role in ensuring the stable operation of systems and improving economic benefits. However, most previous studies ignored limitations a single prediction model useful information error factors, resulting low accuracy. Therefore, this paper proposes multi-stage integrated based on decomposition, multi-objective evolutionary algorithm decomposition (MOEA/D). The proposed consists three stages: first stage, gated recurrent unit (GRU) is used to predict components complete ensemble empirical modal with adaptive noise, new data sets are obtained by combining them original fully mine characteristics. In second MOEA/D angle distance selection strategy population generation optimize GRU network parameters accuracy diversity as objective functions, obtaining several models that consider diversity. third nonlinear integration method optimized integrate values values, considering factors further improve Experimental results Australian wholesale electricity market energy datasets show outperforms comparative terms generalization can be widely applied forecasting.

Language: Английский

---

Electricity consumption prediction based on a dynamic decomposition-denoising-ensemble approach

Engineering Applications of Artificial Intelligence, Journal Year: 2024, Volume and Issue: 133, P. 108521 - 108521

Published: May 9, 2024

Language: Английский

---

MGMI: A novel deep learning model based on short-term thermal load prediction

Applied Energy, Journal Year: 2024, Volume and Issue: 376, P. 124209 - 124209

Published: Aug. 18, 2024

Language: Английский

---