Hybrid Grey Wolf Optimizer-Neural Network Model for Electricity Demand Forecasting in Indonesia

Strategic Insights -

Rapid economic development and energy transition goals in Indonesia necessitate accurate long-term electricity demand forecasting to ensure supply security while optimizing infrastructure investments. Researchers from the Electrical Engineering Department have developed a hybrid Grey Wolf Optimizer-Neural Network (GWO-NN) model to address critical gaps in existing forecasting methodologies. The GWO-NN framework incorporates 15 years of historical data across seven key variables to predict electricity demand in Indonesia from 2026 to 2034.

Key Takeaways:

  • The GWO-NN model outperforms traditional ARIMA models and recent CNN-LSTM approaches in long-term electricity demand forecasting, achieving exceptional 3.9% average absolute difference in conservative scenarios.
  • The model demonstrates significant improvement over traditional ARIMA models (35% error) and recent CNN-LSTM approaches (25% error) in moderate scenarios.
  • The 2034 demand projections range from 377.0 TWh (Conservative) to 546.1 TWh (Optimistic), providing policymakers with robust planning envelopes.
  • The research contributes methodologically through hybrid metaheuristic optimization and practically through evidence-based planning support for Indonesia's renewable energy transition and carbon neutrality targets by 2060.
  • The GWO-NN framework incorporates 15 years of historical data (2010-2025) across seven key variables, including GDP growth, population dynamics, temperature variations, industrial activity, urbanization rates, energy efficiency, and electrification progress.
  • The model's performance is rigorously validated against PLN's official projections.

Statistics:

  • Average absolute difference in conservative scenarios: 3.9%
  • Average absolute difference in moderate scenarios: 19.0%
  • Error rate of traditional ARIMA models: 35%
  • Error rate of recent CNN-LSTM approaches: 25%
  • Projected electricity demand in Conservative scenario (2034): 377.0 TWh
  • Projected electricity demand in Optimistic scenario (2034): 546.1 TWh

Sources:

  • Grey Wolf Optimizer-Neural Network Model for Indonesia Electricity Demand Prediction: Multi-Scenario Analysis and Performance Evaluation 2026-2034. Protek: Jurnal Ilmiah Teknik Elektro, 2025,12(3):156-173.
  • DOI: https://doi-org.sdpl.idm.oclc.org/10.33387/protk.v12i3.10398 (Free version available)
  • Authors: Sofyan Sofyan, Usman Usman, Alamsyah Achmad, Mochammad Apriyadi Hadi Sirad, Ahmad Fudholi, Norazliani MD Sapari