Sign in Subscribe
Machine learning

Machine Learning Model Accurately Forecasts Vietnam's Electricity Consumption

Strategic Insights

— 1 min read

Researchers from Pusan National University have developed a radial basis function neural network (RBFNN) model that accurately forecasts Vietnam's long-term annual electricity consumption. The model, trained on data from 1990 to 2015, was validated using actual consumption data from 2016 to 2020 and was found to outperform the power development plan VIII (PDP VIII) in predicting electricity consumption. The study, published in the journal Energy, provides new insights into the potential of machine learning models in forecasting energy consumption.

Key Takeaways:

  • The RBFNN model was trained on data from 1990 to 2015 and validated using actual consumption data from 2016 to 2020.
  • The study found that the RBFNN model accurately forecasted electricity consumption for the period 2021 to 2023, while PDP VIII overestimated actual consumption.
  • The study concluded that the RBFNN model is more accurate than PDP VIII in forecasting electricity consumption, with an average annual growth rate of 7.2% compared to PDP VIII's 8.7%.
  • The study used the mutual information and SHapley Additive exPlanations (SHAP) method to determine the most appropriate combination of socioeconomic input variables for the optimal forecasting model.
  • The study provides new insights into the potential of machine learning models in forecasting energy consumption and can be applied to similar energy consumption prediction problems.

Statistics:

  • 1990-2015: Training data period for the RBFNN model
  • 2016-2020: Validation data period for the RBFNN model
  • 2021-2023: Forecasting period for the RBFNN model
  • 7.2%: Average annual growth rate of electricity consumption forecasted by the RBFNN model
  • 8.7%: Average annual growth rate of electricity consumption predicted by PDP VIII
  • 2024-2030: Forecasting period for the RBFNN model

Sources:

  • Forecasting Annual Electricity Consumption In Vietnam Using Radial Basis Function Neural Network. Energy, 2025;334.
  • Pusan National University, Dept. of Economics, Busan 46241, South Korea.

Read more