Published Paper Details:

Accurate energy demand forecasting is crucial for optimizing power distribution, reducing energy waste, and improving resource efficiency. Traditional forecasting methods often suffer from limitations in accuracy and scalability. Using a variety of algorithms, such as Support Vector Machine (SVM), Artificial Neural Networks (ANN), k-Nearest Neighbours (kNN), and XGBoost, this study suggests a machine learning-based method to forecast energy consumption. Real-world data from commercial buildings in Malaysia is utilized, considering the influence of environmental factors such as temperature. The dataset is pre-processed to deal with missing values and outliers in order to guarantee dependable model training. The predictive performance of different models is evaluated using metrics such as RMSE, NRMSE, MAE, and MAPE. The results indicate that machine learning models provide significant improvements in forecast accuracy, with ANN and KNN outperforming traditional approaches. This study contributes to energy management by offering an interpretable and scalable solution for demand prediction in smart grids.