What question did this study set out to answer?

The aim is to forecast electricity demand in smart grids using machine learning techniques based on historical data.

April 1, 2026Open Access

Intelligent Forecasting of Power Consumption of Smart Grids with Machine Learning and Full-Stack Implementation on Machine Learning

View Full Paper

Ask AI

Bookmark

Authors

Intelligent forecasting demonstrates machine learning's accuracy in predicting energy demand for smart grids, suggesting enhanced efficiency.

The aim is to forecast electricity demand in smart grids using machine learning techniques based on historical data.
Analyzed a 5-year history of hourly smart grid data for forecasting.
Compared various ensemble models: Random Forest, Extra Trees, Gradient Boosting, LightGBM, and XGBoost.
Used Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) as error indicators for model evaluation.
Developed a full-stack system with Fast API and Streamlit to implement the best model.
XGBoost achieved the lowest MAE of 2236.70 and the lowest RMSE of 3192.55 among the models tested.
The full-stack implementation performed well, with MAE being only 2.7% of the average demand.
Optimization attempts of XGBoost using Optuna were not significantly effective, indicating the default settings were optimal.

Venkat Baba Yemineni (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Intelligent forecasting demonstrates machine learning's accuracy in predicting energy demand for smart grids, suggesting enhanced efficiency.