PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
January 17, 2026MathematicsOpen Access

A Novel Algorithm for Determining the Window Size in Power Load Prediction

View Full Paper
Ask AI
Bookmark
Share

Authors

HLHaobin LiangZSZefang SongYLYiran Liu

Discussion

Loading...

Member takes

Overview

An algorithm optimizes window size for improved power load prediction, indicating enhanced forecasting accuracy across models.

Key Points

  • The aim is to develop an algorithm for scientifically determining the optimal window size in time series forecasting for power load prediction.
  • Proposed a new algorithm based on sample entropy for optimizing window size.
  • Validated the algorithm using open-source Elia grid data across various model architectures.
  • Evaluated performance on recurrent (GRU/LSTM) and attention-based (Transformer) networks.
  • Achieved an optimal window size of 106 for improved predictions.
  • CEEMD-GRU model showed a MAPE of 0.256, RMSE of 22.529, and MAE of 18.186.
  • Reduced prediction errors by over 5% compared to the undecomposed benchmark, particularly enhancing decomposed sequences.

Cite This Study

Liang et al. (2026) studied this question.

synapsesocial.com/papers/696b2672d2a12237a9349b5fhttps://doi.org/10.3390/math14020304
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Long Short-Term Memory1997 · 102,018 citations
  2. 2Wind power forecasting based on outlier smooth transition autoregressive GARCH model2016 · 38 citations
  3. 3The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis1998 · 23,737 citations
  4. 4Asymmetric GARCH type models for asymmetric volatility characteristics analysis and wind power forecasting2019 · 44 citations
  5. 5Short-Term Forecasting-Based Network Reconfiguration for Unbalanced Distribution Systems With Distributed Generators2019 · 38 citations