Experimental Investigation on Assessment and Prediction of Specific Energy in Rock Cutting

Abstract

The rock cutting machine was fabricated to measure the cutting rate and specific energy (SE). The variable parameters include attack angle, pick angle, RPM, cutting force, and torque to determine the cutting parameters. For measuring the cutting force and torque, a cutting tool dynamometer is used. Experimental investigations were also carried out to determine physico-mechanical properties of the rocks, namely density, uniaxial compressive strength (UCS), Brazilian Tensile strength (BTS), abrasivity and brittleness of the rocks were determined as per ISRM standards. During the cutting process, the RPM is varied from 225, 300,325 and 350 and the cutting force is measured at each RPM. The cutting process was carried using point attack picks of 45°, 50°, 55° and 65° pick angles and 45°, 55° and 65° attack angles. During the cutting process, the cutting force was varied using a hydraulic pressure valve. In this research, for each RPM and thrust combination, cutting is done for 60 seconds and cutting depth is measured using a digital vernier calliper. The rock cuttings are collected and weighed using a digital weighing machine. Then, the SE (J/m3) is calculated by cutting force multiplied by the depth of cut and divided by volume collected during the cutting process. The increase in RPM, torque, and cutting force observations reveals that the increase in the parameters increases the cutting rate with a corresponding decrease in SE. With cutting rate, the minimum and maximum variation irrespective of the rock type are found to be 0.3 to 4.8% for pick angles, 0.2 to 32% for attack angles, 0.05 to 4.08% for RPM, 0.05 to 3.2% for torque and 0.05 to 3.2% for cutting force. With specific energy, the minimum and maximum variations irrespective of the rock type are found to be 0.023 to 4.41% for pick angles, 21.91 to 51.26% for attack angles, 0.03 to 4.41% for RPM, 0.03 to 7.8% for torque and 0.18 to 7.36% for cutting force. Hence, attack angle has more influence on cutting rate and specific energy. The cutting rates and specific energy values were determined for the pick tool subjected to wear of 5mm at an 45° attack angle. a comparison of the same was made. A decrease in cutting rate is observed with a proportional increase in specific energy. The minimum vi and maximum variations irrespective of the rock type are 24.5 to 33.36% for pick angles, 24.5 to 30.36% for RPM, 21.56 to 35.16% for torque and 20.05 to 32.61% with cutting force for cutting rate. For specific energy, the minimum and maximum variations irrespective of the rock type are 21.86 to 35.81% for pick angles, 21.80 to 32.66% for RPM, 21.89 to 36.20% for RPM torque and 21.98 to 36.64% for cutting force. A property correlation with specific energy was also plotted as a line graph It was observed that, with the increase in density, UCS, BTS, abrasivity, and brittleness of the rock, SE increases linearly. This is because, with the increase in the strength of rock, the cutting resistance increases linearly. The regression models shown in Equations 6.1, 6.2 and 6.3 were developed and can be used to estimate the SE during rock cutting as they can be used as guidance in practical applications. The developed regression model results showed that the SE's significant operating variables were attack angle, type of pick followed by other cutting parameters, such as the rock's mechanical properties. The results showed that input parameters were significant, and the model possesses an R-Square value of 99.55%. The respective variance account for (VAF), root mean square error RMSE and mean absolute percentage error (MAPE) indices for predicting SE are VAF of 99.17, RMSE of 12.08 and MAPE of 0.032535, respectively, from the multiple regression model (testing). The result of the current study provides opportunities to evaluate the cuttability of rocks before involving complicated experimental procedures. Error graphs also resulted in the goodness of fits of a statistical model. Artificial Neural Network (ANN), was developed to predict the SE. the input parameters include cutting force, pick angle, attack angle, depth of cut, volume broken and rock properties like density, UCS, BTS, abrasivity and brittleness. The ANN results showed that the model's predictive performance for VAF, RMSE and MAPE indices are VAF of 99.98289, RMSE of 9.47741, MAPE of 0.0000158 for training and VAF, RMSE and MAPE for validation were VAF of 99.97602, RMSE of 11.85352, MAPE of 0.0000666. Error graphs also resulted in the goodness of fits of a statistical model. vii A numerical model using Finite Element Method (FEM) analysis was constructed to determine the depth of cut for all pick-rock combinations considered using the cutting force values from experimental rock cutting tests (up to loading cycle only). Then the depth of penetration obtained in FEM analysis of all pick-rock combinations was compared with the respective depth of cut obtained with experimental results. The depth of penetration obtained during experiments is lesser than FEM analysis for all pick-rock combinations considered and ranges from 1 to 8% (except a few). Further, the results indicated that displacement decreases from the loading axes towards the boundary in all directions. The stress analysis was carried using Ansys workbench for all the pick-rocks combinations considered along X, Y and Z - directions. The results showed that the maximum compressive stress generated is at the tip of the cut zone. In this research, a new concept is proposed: Rock Cutting Resistance (RCR), i.e., the resistance offered by the rock against the cutting force required to achieve a unit depth of cut, and is expressed as N/mm. The results of the RCR (Experimental and FEM) can be used to predict the depth of cut during rock cutting. Hence, RCR can be used for the efficient design of the rock cutting parameters and the machine.