Experimental Investigations on Assessment and Prediction of Specific Energy in Rock Indentation Tests

Abstract

Indentation is a fundamental process in drilling and cutting/sawing of rocks. Assessment of specific energy (SE), which is energy required to excavate (drilling or cutting) a unit volume of rock is important because it is a one of the parameters to determine drillability and cuttability of rocks. Drillability of rocks is an important parameters to decide the progress and drilling costs of the excavation. Similarly index angle, which is amount of rotation of bit between successive blows, play an important role in percussive drilling. Static indentation tests were carried out in six types of rocks viz. marble, limestone, basalt, steel gray granite, moon white granite and black galaxy granite by using commercial drill bits of Chisel, Cross and Spherical button of 35mm, 38mm, 45 mm and 48mm diameters as indenters on Micro controller compressive testing machine. The loading was done on rock specimen considered for 60 seconds, and then unloading was done. During loading and unloading, at every 5 seconds forces and penetrations were recorded. Then, F-P curves were drawn for all bit-rock combinations considered. The volume of rock excavated was determined using the density of the rocks. Then, SE (ratio of area under F-P curve (energy expended) to the volume of rock excavated)) was calculated for all bit-rock combinations considered. These experimental investigations were carried out in the laboratory with an objective to determine the SE during indentation and to study the influence of index angle on specific energy. The graphs were drawn between index angle and SE. The results showed that the SE is minimum at 30° index angle for the rocks like marble, limestone whereas; the SE is minimum at 20° index angle for the rocks like basalt, steel gray granite, moon white granite and black galaxy granite. Experimental investigations were also carried out to determine physicomechanical properties like density, uniaxial compressive strength (UCS), Brazilian Tensile strength (BTS), Hardness (Schmidt Rebound Number (SRN)), Young’s Modulus, Poisson’s ratio. Regression analysis was carried between SE and above properties to study the influence of the physico-mechanical properties on SE. It was observed that, with the increase in density, UCS, BTS, Hardness (SRN), Young’s Modulus, Poisson’s ratio of the rock, SE increases. This is because of the fact that, with the increase in thevi strength, the resistance to indentation increases. But with increase in percentage abrasion resistance, SE decreases. Similarly the thin section analysis, petrographic studies and XRay Fluorescence tests were carried to find the mineralogical composition of rocks considered and regression analysis between minerals present in all rocks considered and SE to find the influence of mineralogical composition on SE. The results showed that except in the case of feldspar, an increase in SE, as the percentages of quartz, hornblende, pyrites, magnetite and biotite mica increases. Similarly predictive models (Regression analysis and Artificial Neural Network (ANN)) were developed to predict the specific energy from operating variables like diameter of bit and index angle and selected properties of rocks like density, UCS, BTS, abrasion resistance, Hardness (SRN), Young’s Modulus, Poisson’s ratio. The results showed that operating variables and above properties of rocks are significant parameters to predict the specific energy. Further, results (for spherical button bit -prediction performance indices (VAF: 90.18(regression), 99.05135(ANN), RMSE: 6.58(regression), 2.16(ANN), and MAPE 0.19(regression), 0.055 (ANN))) showed that the predictive performance of ANN model are higher than those of multiple regression equations. So, ANN is a good approach for minimizing the uncertainties in the rock and soil engineering projects. The Numerical Modelling (Finite Element Method analysis) was carried out to determine the depth of penetration for all bit-rock combinations considered by using the force values from static indentation test (up to loading cycle only). Then the penetration obtained in FEM analysis of all bit-rock combinations considered were compared with experimental results. The numerical value indicates that experimental values are higher than FEM analysis and ranges from 10 to 19.5% (except few). Further the results indicated that in all the directions, displacement is decreasing from the loading axes towards the boundary. The stress analysis also was carried in all the bit-rocks combinations considered along X- Y and Z- axes. The results showed that maximum compressive stress is generated near the tip of the bit and the magnitude of compressive stresses developed at any point away from vertical axis depends on the geometry of the indenter.