A novel hybrid wind speed prediction framework based on multi-strategy improved optimizer and new data pre-processing system with feedback mechanism

Zhirui Tian and Mei Gai

Energy, 2023, vol. 281, issue C

Abstract: As a kind of renewable energy, wind energy has great potential for development and has been paid attention to by governments all over the world. However, due to the high uncertainty of wind speed, how to accurately predict wind speed and make use of wind energy has been recognized as a difficult problem. In order to solve this problem, a new hybrid wind speed prediction framework is proposed, which is composed of two subsystems, data preprocessing system and high-accuracy prediction system. In the system 1, the feedback mechanism is creatively added to the singular spectrum analysis (SSA) to find out the optimal decomposition-recombination strategy through the accuracy feedback. In the system 2, the unconstrained weighting mechanism is realized through the combination of combined neural network and multi-objective optimization algorithm to maximize the prediction accuracy on the premise of ensuring the stability of prediction. Besides, an improved meta-heuristic optimization algorithm based on cross-perturbation strategy (CP-JAYA) and its multi-objective form (MO-CPJAYA) are applied on two systems respectively to further improve the prediction ability of the framework. In 5 groups of experiments, the accuracy, advancement, generalization and sensitivity of the model are tested and compared with 13 other models. The proposed prediction framework has the best performance in all four sets of data. In 3 groups of discussions, we verify the advanced nature of CP-JAYA and MO-CPJAYA respectively through 13 single-objective test functions (CEC) and 4 multi-objective test functions (ZDT), and the speed advantage of the framework by recording the CPU running time.

Keywords: Wind speed prediction; Singular spectrum analysis; Feedback mechanism; Multi-strategy improved optimizer; Unconstrained weighting (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544223016195
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:281:y:2023:i:c:s0360544223016195

DOI: 10.1016/j.energy.2023.128225

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().