Stability Analysis of Rock Structure in Large Slopes and Open-Pit Mine: Numerical and Experimental Fault Modeling

摘要

Deep open-pit mines and large rock slopes expose many diverse rock lithologies and geological structures (e.g., faults, bedding planes) that may reduce the integrity of slopes. Numerical modeling is a powerful tool for simulating these structures;;however, there are few guidelines and methods for calibrating/validating and implementing faults in a numerical model. This paper presents a novel laboratory method to calibrate numerical models and highlights the challenges in simulating faults. One of the main issues in reliable modeling of faulted rock structure is the scarcity of experimental analyses in the laboratory under the controlled conditions. Moreover, a comprehensive evaluation of the effect of using the conventional fault modeling methods on the stability of rock structures is required, as well as a benchmarking between theoretical and experimental results. This research combines theory and experiment, to fill the existing gaps, using numerical simulation and laboratory measurements. Using FLAC3D software, the sensitivity and comparative analyses are carried out for the numerical simulations to investigate the stability of rock slopes on large and small scales (overall open-pit slope and bench slope), and the fault zones. The weak zone (WZ), ubiquitous-joint (UJ), and interface (IF) techniques are the widely used methods in the modeling to capture fault slip mechanisms. The factor of safety (FOS) of the slope is monitored upon variation of the design parameters, such as fault and rock mass mechanical properties, fault types, and modeling framework (e.g., mesh density, convergence ratio). In addition, parameters such as shear displacement and shear stress are investigated to deduce the failure mechanism of the studied models. Finally, laboratory tests were performed to calibrate the modeling results and approximate the agreement between theoretical and experimental results. The results of sensitivity analysis showed that choosing an adequately low convergence ratio is critical for estimating FOS. However, beyond a certain convergence ratio, below 10-7, this change is negligible (less than 5%). The results of mesh density sensitivity analysis indicate that the FOS values are insensitive to the mesh density in the WZ method (less than 5% change in FOS), the IF method shows the median sensitivity (5-12% change in FOS), and the UJ method is the most sensitive (FOS values improves by -31%). Comparison between laboratory test and numerical modeling (FOS_(lab) =1.71, FOS_(wz) =1.51, FOS_(IF) = 1.62, and FOS_(UJ)= 1.76) indicates a good agreement between the UJ and IF methods and the laboratory model (~ 3-5% discrepancy). It needs to be mentioned that these analyses/tests are not to favor one method over the other, but rather to emphasize the pros and cons of each within the assumptions of this study.