Cited by Approach for Short-Term Power Load Prediction Utilizing the ICEEMDAN–LSTM–TCN

A day-ahead industrial load forecasting model using load change rate features and combining FA-ELM and the AdaBoost algorithm DOI

Ziwei Zhu, Mengran Zhou, Feng Hu

et al.

Energy Reports, Journal Year: 2022, Volume and Issue: 9, P. 971 - 981

Published: Dec. 19, 2022

Industrial customers consume a large part of the total electricity demand. In operation industrial energy systems, accurate prediction electric loads is prerequisite to help users adjust their load dispatch and improve efficiency. Therefore, this paper proposes day-ahead forecasting model employing change rate features combining firefly algorithm optimize extreme learning machine adaptive boosting (LCR-AdaBoost-FA-ELM). The mainly influenced by power users' production schedules, making its laws analyzed changing data itself. Given this, feature introduced form candidate set with variables such as date lag load. order decrease number parameters required train model, Spearman correlation coefficient used select high-quality input eliminate that are weakly associated consumption. basic ELM, based on which FA weights biases. Finally, ensemble concept learn combine multiple FA-ELM weak predictors AdaBoost correct errors. paper, proposed validated using typical industry, furniture factory, research case. results show LCR can capture nonlinear characteristics sequence, resulting in more precise outcomes. Additionally, ELM accuracy, lower error once more. Using mean absolute percentage (MAPE) an example, AdaBoost-FA-ELM declines 76.85% compared decreases 23.90% before after applied. framework provides new strategy for field forecasting.

Language: Английский

Citations

Multivariate solar power time series forecasting using multilevel data fusion and deep neural networks DOI

Sarah Almaghrabi, Mashud Rana, Margaret Hamilton

et al.

Information Fusion, Journal Year: 2023, Volume and Issue: 104, P. 102180 - 102180

Published: Dec. 9, 2023

Accurate forecasting of regional solar photovoltaic power (SPVP) generation is essential for efficient energy management and planning. Existing approaches have shown the effectiveness decomposing time series to model stochastic variability in SPVP data. However, these limitations extracting exploiting both spatial temporal information from complex high-dimensional data multiple sources with intricate relationships, which can impact accuracy predictions. In this paper, we propose a novel approach called multilevel fusion neural basis expansion analysis (MF-NBEA) aggregated regional-level generation. MF-NBEA integrates exogenous at levels, uses supervised unsupervised encoders provide compact representation, enhances learning by incorporating information. It also includes sequence analyser module based on network decomposition mechanism learn incorporates residuals learner improve overall We evaluate using two real-world datasets find that it outperforms state-of-the-art deep methods terms forecast accuracy. Furthermore, facilitates knowledge extraction interpretable predictions regarding trend, seasonality, residual components. The insights gained our inform decision-making planning, lead more sustainable resource utilisation.

Language: Английский

Citations

A synchronized multi-step wind speed prediction with adaptive features and parameters selection: Insights from an interaction model DOI

Wenxin Xia, Jinxing Che, Kun Hu

et al.

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: 255, P. 124764 - 124764

Published: July 14, 2024

Language: Английский

Citations

Short-term electrical load forecasting based on pattern label vector generation DOI

Haozhe Zhu,

Qingcheng Lin,

Xuefeng Li

et al.

Energy and Buildings, Journal Year: 2025, Volume and Issue: unknown, P. 115383 - 115383

Published: Jan. 1, 2025

Language: Английский

Citations

Twin extreme learning machine model and cooperation search algorithm for multi-step-ahead point and interval runoff prediction DOI

Zhong-kai Feng, Pan Liu, Wen-jing Niu

et al.

Journal of Hydrology, Journal Year: 2025, Volume and Issue: unknown, P. 132778 - 132778

Published: Jan. 1, 2025

Language: Английский

Citations

A hybrid system with optimized decomposition on random deep learning model for crude oil futures forecasting DOI

Jie Wang, Ying Zhang

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 126706 - 126706

Published: Feb. 1, 2025

Language: Английский

Citations

A multi-modal feature combination mechanism for identification of harmonic load in distribution networks based on artificial intelligence models DOI

Renzeng Yang, Shuang Peng, Gang Yao

et al.

International Journal of Electrical Power & Energy Systems, Journal Year: 2025, Volume and Issue: 166, P. 110519 - 110519

Published: Feb. 14, 2025

Language: Английский

Citations

Enhanced forecasting method for realized volatility of energy futures prices: A secondary decomposition-based deep learning model DOI

Hao Gong, H. Y. Xing,

Qianwen Wang

et al.

Engineering Applications of Artificial Intelligence, Journal Year: 2025, Volume and Issue: 146, P. 110321 - 110321

Published: Feb. 20, 2025

Language: Английский

Citations

Short-Term Electrical Load Forecasting Using an Enhanced Extreme Learning Machine Based on the Improved Dwarf Mongoose Optimization Algorithm DOI

Haocheng Wang, Zhang Yu,

Lixin Mu

et al.

Symmetry, Journal Year: 2024, Volume and Issue: 16(5), P. 628 - 628

Published: May 18, 2024

Accurate short-term electrical load forecasting is crucial for the stable operation of power systems. Given nonlinear, periodic, and rapidly changing characteristics forecasts, this paper introduces a novel method employing an Extreme Learning Machine (ELM) enhanced by improved Dwarf Mongoose Optimization Algorithm (Local escape Algorithm, LDMOA). This addresses significant prediction errors conventional ELM models enhances accuracy. The enhancements to include three key modifications: initially, dynamic backward learning strategy integrated at early stages algorithm augment its global search capabilities. Subsequently, cosine employed locate new food sources, thereby expanding scope avoiding local optima. Lastly, “madness factor” added when identifying sleeping burrows further widen area effectively circumvent Comparative analyses using benchmark functions demonstrate algorithm’s superior convergence stability. In study, LDMOA optimizes weights thresholds establish LDMOA-ELM model. Experimental forecasts utilizing data from China’s 2016 “The Electrician Mathematical Contest in Modeling” that model significantly outperforms original terms error

Language: Английский

Citations

Research on semi-supervised soft sensor modeling method for sulfur recovery unit based on ISSA-VMD-ESN DOI

Qinghong Wang,

Longhao Li

Chemical Engineering Science, Journal Year: 2024, Volume and Issue: 298, P. 120397 - 120397

Published: June 19, 2024

Language: Английский

Citations