Energy Consumption Prediction in the Steel Industry Using Principal Component Analysis and Regression Tree Methods

The steel industry is a major contributor to global energy consumption and greenhouse gas emissions, driving the need for enhanced efficiency. This study explore the use of Principal Component Analysis (PCA) combined with Regression tree (RT) techniques to predict energy consumption in the steel industry. PCA reduces data dimensionality, while RT addresses complex, non-linear relationships. Tested on a Kaggle dataset, the PCA and RT model achieved high accuracy, with a Root Mean Square Error (RMSE) of 0.67 and an accuracy rate of 90.82%, outperforming other methods in a comparative analysis. The model’s moderate training time of 1.18 seconds highlights its efficiency. Visual and comparative analysis confirmed the model’s strong alignment with observed energy consumption values and its balance between accuracy and computational efficiency. The PCA and RT model is an effective tool for predicting energy consumption in the steel industry, offering a practical approach to improving energy efficiency and sustainability. Future research could explore advanced techniques to further enhance predictive accuracy and model robustness.