A Hybrid Numerical–Evolutionary Framework for Optimization of Underground Blast Design
Underground coal mine blasting is often associated with excessive flyrock, overbreak, and high explosive consumption, primarily due to inefficient distribution of blast energy in confined environments. This study develops a coupled LS-DYNA–Genetic Algorithm (GA) optimization framework for underground blasting based on the principle of controlled energy transfer. The reported performance indicators should be interpreted within the operational and geomechanical constraints of the studied coal mine, where conservative blasting practice governs the achievable advance and excavation response. The geomechanical properties of the Hojedk coal mine were characterized and incorporated into a three-dimensional explicit dynamic model to simulate stress-wave propagation, damage evolution, and fragmentation. A GA was subsequently employed to optimize key blast design parameters, including burden, spacing, charge distribution, and a normalized energy-consumption index, under multi-objective constraints aimed at minimizing blast-induced damage while maximizing excavation efficiency. The optimized blast pattern demonstrates a significant improvement in performance compared to the existing design, with a 19% increase in advance per cycle, a 25.6% reduction in explosive consumption per round, a 37.4% reduction in physical powder factor, a decrease in flyrock velocity of up to a 63.5% reduction in flyrock velocity, and an approximately 50% reduction in overbreak. Numerical results indicate that the improved performance is primarily attributed to enhanced energy redistribution toward the free face and effective stress-wave attenuation through the incorporation of relief mechanisms. The proposed coupled numerical–evolutionary approach provides a robust and transferable methodology for optimizing underground blast design, offering substantial benefits in terms of safety, energy efficiency, and operational cost reduction in underground mining environments. The framework is therefore particularly suited for constrained underground environments where safety and stability govern blasting design decisions.