EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Long-term electricity load forecasting model based on multi-period and multi-scale analysis

Xiaolin Zhang, Zhi Liu iD, Chunyang Wang, Zhen Qian

PLOS One · 30.10.2026 · 21 (10) · e0361106

DOI: 10.1371/journal.pone.0361106 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 30.10.2026. Verfasst von Xiaolin Zhang und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Electronic Information Engineering, Changchun University of Science and Technology, Changchun, Jilin, China. Hauptthemen laut MEDLINE: Electricity, Models, Theoretical. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Existing long-term power load forecasting models face significant challenges due to the complexity of non-stationary electricity data, inadequate information representation, and suboptimal prediction accuracy. To address these issues, this study introduces PowerNet, a novel forecasting framework based on a multi-period and multi-scale approach. PowerNet first employs Fast Fourier Transform (FFT) to perform multi-period analysis, transforming one-dimensional time series data into two-dimensional (2D) tensors to capture latent periodic patterns. A multi-scale information representation module is then implemented using computer vision-inspired techniques, including Convolutional Neural Networks (CNN), Gated Recurrent Units (GRU), and non-local attention mechanisms to extract robust features from these tensors. Finally, a fully connected layer performs feature reconstruction, while residual connections facilitate multi-step load forecasting. PowerNet was evaluated against state-of-the-art models such as Informer, LogTrans, and ETSformer across two public electricity datasets and one weather dataset. At prediction horizons of 96, 192, 336, and 720 time steps, experimental results demonstrate that PowerNet achieves competitive or superior accuracy on the evaluated datasets. Notably, compared with Autoformer, PowerNet reduces the Mean Absolute Error (MAE) by 18.77% to 37.41% and the Mean Square Error (MSE) by 18.23% to 47.74%, significantly enhancing the reliability of long-term electricity load forecasting.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ElectricityModels, TheoreticalForecastingPrediction AlgorithmsConvolutional Neural NetworksRecurrent Neural NetworksFourier AnalysisPredictive Learning Models

Autor:innen und Institutionen

  • School of Electronic Information Engineering, Changchun University of Science and Technology, Changchun, Jilin, China.
  • Yahe School of Design and Engineering, Haikou University of Economics, Haikou, Hainan, China.
  • Tsinghua University, Beijing, China.
  • Beijing Jiaotong University, Beijing, China.

ORCID-Profile: Zhi Liu

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Zhang, X., Liu, Z., Wang, C., & Qian, Z. (2026). Long-term electricity load forecasting model based on multi-period and multi-scale analysis. PLOS One, 21(10), e0361106. https://doi.org/10.1371/journal.pone.0361106

Vancouver

Zhang X, Liu Z, Wang C, Qian Z. Long-term electricity load forecasting model based on multi-period and multi-scale analysis. PLOS One. 2026;21(10):e0361106. doi:10.1371/journal.pone.0361106

BibTeX
@article{Zhang2026,
  author = {Zhang, Xiaolin and Liu, Zhi and Wang, Chunyang and Qian, Zhen},
  title = {Long-term electricity load forecasting model based on multi-period and multi-scale analysis},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {10},
  pages = {e0361106},
  doi = {10.1371/journal.pone.0361106}
}
Kostenlos starten →