Image processing technique to find the burden rock velocity and its interdependency on bench blasting parameters – a case study

In any rock engineering or excavation project, optimizing blast design is essential for controlling post-blast phenomena such as flyrock, burden throw, and burden rock velocity (BRV). This study investigates the dynamic response of BRV in bench blasting operations by employing a high-speed camera and ProAnalyst software to quantify rock displacement. The relationships between BRV and key blast parameters, number of holes (NH), explosive charge per hole (EPH), total explosive charge (TEC), stiffness ratio (K), and stemming ratio (STR), were analyzed to identify linear or non-linear trends. Non-linear trends, including power-law relationships for NH and TEC, exponential for EPH, logarithmic for STR, and quadratic for K, dominate the parametric interactions with burden rock velocity, as evidenced by high R² values. Predictive models using multiple linear regression (MLR) and multiple non-linear regression (MNLR) were developed, with MNLR demonstrating higher accuracy (R² = 0.959) due to its ability to capture non-linear interactions between bench blasting parameters and burden rock velocity. Features importance analysis based on the cosine amplitude method highlights STR, TEC and NH as the most influential factors. This study underscores the significance of high-speed cameras in assessing rock burden movement in terms of BRV, a crucial parameter in controlled bench blasting. The findings offer an indirect assessment of operational efficiency in mining by mitigating excessive rock movement and associated hazards such as flyrock and ground vibrations.