Energy Consumption Reduction in Underground Mine Ventilation System: An Integrated Approach Using Mathematical and Machine Learning Models Toward Sustainable Mining

This study presents an integrated approach combining the Hardy Cross method and a gradient boosting (GB) optimization model to enhance ventilation systems in underground mines, with a specific application at the Jabal Sayid mine in Saudi Arabia. The Hardy Cross method addresses variations in airflow resistance caused by obstacles within ventilation pathways, enabling accurate predictions of the flow distribution across the network. The GB model complements this by optimizing fan placement, pressure control, and airflow intensity to achieve reduced energy consumption and improved efficiency. The results demonstrate significant improvements in fan efficiency, optimized energy usage, and enhanced ventilation effectiveness, achieving a 31.24% reduction in electricity consumption. This study bridges deterministic and machine learning methodologies, offering a novel framework for the real-time optimization of underground mine ventilation systems. By combining the Hardy Cross method with GB, the proposed approach outperforms traditional techniques in predicting and optimizing airflow distribution under dynamic conditions.