Investigation of discrete element and bonded particle methods for modelling rock mechanics subjected to standard tests and drilling

A structural optimization task has been performed to propose a draft for a new deep drilling system by studying experimental, numerical and theoretical models of the rock, indenter and interaction between rock and drill bit. In order to do so, on the drill side the conventional drilling challenges were recapped. It has been tried to address some of those issues by modifying a self-assisting biomimetic drilling mechanism using dual reciprocating percussive action (known as Dual Reciprocating Drill (DRD)). On the rock side, basics of rock mechanics were reviewed and discussed. Important bedrocks for drilling were targeted with Berea sandstone data and laboratorial testing of intact samples of Pietra Serena sandstone. Through four distinct experimental sessions of standard rock tests, Unconfined/Confined Compression, Brazilian Split (Indirect tensile test) and Four Point Bending Flexural, using various specimen dimensions, some failure envelopes of interest have been approximated. In addition, computational simulations of such tests were planned and performed using hybrid FEM-DEM/BPM approach. This numerical model was calibrated to be exploited in the next step where Punch Penetration test and DRD were targeted as the main design initiator simulation. The result of identical scenarios, performed by other numerical methods (FEM and SPH), were sat against the ones of this study in terms of accuracy, precision and efficiency for sake of decision making on the best method. Abaqus and LS-DYNA were investigated comprehensively through sensitivity analyses and different calibration approaches were offered in this study. Through a series of hybrid FEM-DEM models the interaction of rock fragments and drill head was also simulated accounting for post-failure state of substrate. Those fragment models were calibrated using true triaxial test of coarse sands. Such multiphase analyses of the sandstone percussion allowed the drill action prediction leading to the final design proposal. Optimization of the drill bit and its immediate string promised WOB and mass reduction, ROP increase while lessening the energy consumption. The proposed drill here was foreseen to be mounted on a semi-autonomous ROV (Remotely Operated (underwater) Vehicles) benefited from self-assisting penetration mechanism.

N/A