Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
April 24, 2026International Transactions on Electrical Energy SystemsOpen Access

A Hybrid CNN–BiLSTM–GRU Model for Electric Energy Consumption Forecasting

View Full Paper
Ask AI
Bookmark
Share

Authors

LLLuciano Roberto da Silva LealBBByron Leite Dantas BezerraUniversidade de PernambucoJOJoão Fausto Lorenzato de OliveiraUniversity of Coimbra

Discussion

Loading...

Member takes

Implication

Demonstrates improved accuracy in electricity consumption forecasting using a novel model, indicating better energy management solutions.

Key Points

  • The central aim is to improve forecasting accuracy for electricity consumption in diverse regions.
  • Developed a hybrid CNN–BiLSTM–GRU model for forecasting electricity consumption.
  • Compared model performance with traditional ARIMA and machine learning methods including random forest and support vector regression.
  • Used historical data from the National System Operator (ONS) for model training and validation.
  • CNN–BiLSTM–GRU achieved the lowest mean absolute error (MAE) across all regions analyzed.
  • North region MAE: 0.0368; Northeast: 0.0348; South: 0.0503; Southeast: 0.0336.
  • Statistical validation confirmed the model's robustness against regional disparities, outperforming traditional methods.

Cite This Study

Leal et al. (2026) studied this question.

synapsesocial.com/papers/69eb0a94553a5433e34b48d3https://doi.org/10.1155/etep/4831304
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1A CNN-LSTM-GRU Hybrid Model for Spatiotemporal Highway Traffic Flow Prediction2025 · 13 citations
  2. 2Time series classification from scratch with deep neural networks: A strong baseline2017 · 2,116 citations
  3. 3An ensemble CNN-LSTM and GRU adaptive weighting model based improved sparrow search algorithm for predicting runoff using historical meteorological and runoff data as input2023 · 204 citations
  4. 4Deep Learning for Household Load Forecasting—A Novel Pooling Deep RNN2017 · 1,121 citations