Document Type : Original Research Paper
Authors
1 VNIT, NAGPUR
2 Mining Engineering, Associate Professor, VNIT, Nagpur, India
3 Underground Mining Division, CMPDI, Ranchi, India
4 NaCCER, CMPDI, Ranchi, India
Abstract
Subsidence associated with underground coal mining is a significant geotechnical concern in many coal-producing regions. The extraction of coal over large areas from underground often leads to the collapse of overlying strata into the goaf, subsequently causing surface subsidence. The extent of this subsidence varies widely across mines, depending on several factors, including mine geometry, geological discontinuities, physico-mechanical properties of the overlying strata, extraction method, seam thickness, and depth of working. Among these, the angle of draw (AoD) plays a critical role in delineating the subsidence-affected zone, particularly in underground coal mining. Accurate prediction of AoD is essential for safe mine planning and the mitigation of subsidence-related hazards. In the present study, a comprehensive field investigation was conducted to collect mine operational parameters from various underground coal mines. Using this dataset, Genetic Programming (GP) was employed to model the relationship between AoD and key mining and geological parameters. The developed GP model demonstrated a strong correlation between predicted and measured AoD values, with a coefficient of determination (R) = 0.7921, highlighting the model’s predictive capability. Additionally, a sensitivity analysis (SA) was performed to identify the most influential input parameters affecting AoD. The analysis indicated that, while all five input variables significantly impact AoD, the compressive strength of overlying strata exhibited the highest influence (sensitivity score = 0.98). The findings of this study provide a data-driven approach to predict the angle of draw in underground coal mines, offering valuable insights for improved mine design, extraction strategies, and surface infrastructure protection.
Keywords
Main Subjects
)