Experimental Investigations on Assessment and Prediction of Temperature during Rotary Drilling

Abstract

Drilling is one of the most energy consuming technology processes in conducting exploration works in mining industries. Heat is generated during rock drilling due to friction between rock and bit, which leads to the thermal stress and is subjected to rock failure. Nearly 80% of the energy supplied to the bit is used for heat release, 1.5 % to 10 % for the residual changes of the bit and 8 to 10 % for destruction of rocks (Dreus et al. 2016). Overall, the major cause of the wear of the bit is abrasion wear because the drill bit is most abraded against the rock formation, it is mainly consists of the silica content in the rock samples (Abbas, K. 2018). Hence, selection of drilling parameters and choosing the most appropriate type of drill bit for a certain rock sample will prolong the bit life and reduce the drilling costs. Experimental investigations were carried out on five types of rock samples to measure the temperature at bit-rock interface using newly fabricated K-type grounded thermocouple during rotary drilling in a Computer Numerical Control (CNC) vertical machining centre. The temperature at bit-rock interface was measured by using digital temperature indicator and embedded thermocouple technique. The observations were made using different operational parameters, namely, drill bit diameter (6, 8, 10, 12 and 16 mm), spindle speed (250, 300, 350, 400 and 450 rpm), penetration rate (2, 4, 6, 8 and 10 mm/min) and at different depth (6, 14, 22 and 30 mm). The experimental results show that the tungsten carbide masonry drill bit for all bit rock combinations considered generates an average maximum temperature at bit-rock interface of 91C, 128C, 236C, 124C and 147C for medium grained sandstone, fine grained sandstone (grey), fine grained sandstone (pink), limestone and shale respectively. It was found that the temperature at bit-rock interface increased significantly from 48.48C to 74.53C, 43.81C to 84.68C, 55.73C to 147.82C, 50.34C and 44.29C to 89.57C for MG, FG, FGP, limestone and shale with the increase in depth of drill, drill bit diameter, spindle speed and penetration rate.vi Wear rate of the tungsten carbide (WC) drill bit was measured using weight loss method. Wear rate of tungsten carbide (WC) drill bit and the interrelationship between temperature at the bit-rock interface and wear rate during rotary drilling operations was investigated. Under the test conditions at constant drill bit diameter (16mm) and spindle speed (450rpm) by varying penetration rate of 2, 4, 6, 8 and 10 mm/min respectively, wear rate coefficients (k) were calculated using Archard model and found to be 0.05916, 0.0705, 0.07924, 0.05423 and 0.0596 mg/Nm for medium grained sandstone, fine grained sandstone (grey), fine-grained sandstone (pink), limestone and shale respectively. The SEM and EDS analyses clearly showed that medium grained (MG) sandstone and limestone have less SiO2 contents of 7.85 (wt.%) and 5.76 (wt. %), which gave less wear rate coefficient and bit-rock interface temperature of 0.1423, 0.1594 mg/Nm and 74C, 83C respectively. Similarly fine grained sandstone (grey), fine grained sandstone (pink) and shale which contain 16.45 (wt. %), 30.22 (wt. %) and 25.54 (wt. %) gave wear rate coefficients and temperatures at bit-rock interface of 0.2072, 0.2803, 0.1781 mg/Nm and 84C, 147C, 89C respectively. A study of rock drilling was systematically carried out using full factorial method to determine the effect of operational parameters such as drill bit diameter, spindle speed, penetration rate, thrust and torque on the temperature at bit-rock interface and to find out the wear rate of the tungsten carbide masonry drill bit. ANOVA analysis was applied to investigate the effect of the operational parameters (drill bit diameter, spindle speed, penetration rate, thrust and torque) and rock properties (uniaxial compressive strength, Brazilian tensile strength, Los Angeles abrasion and density) on temperature at bit-rock interface in rotary drilling. The statistically significant parameters i.e., operational parameters and rock properties were identified. Furthermore, multiple linear regression analysis and artificial neural network (ANN) technique were utilized to develop empirical models for predicting the temperature at bit-rock interface and wear rate of the tungsten carbide drill bit. The developed models showed good predictive capability with acceptable accuracy. Artificial neural network models showed optimum neuron with best algorithm and transfer function for the prediction of temperature at bit-rock interface.