Investigation on the relationship among the Cerchar abrasivity index, drilling parameters and physical and mechanical properties of the rock

Abstract

Rock abrasivity is an important factor that affects tool wear and tool parameter design, construction efficiency, and cost budgeting during rock excavation. The Cerchar abrasivity index (CAI) is used as a standard parameter to characterize the abrasiveness of rocks. It can be obtained through the Cerchar abrasivity test in a laboratory. However, conventional laboratory Cerchar abrasivity tests and petrophysical and mechanical tests are associated with certain problems, such as long cycles, high costs, and measurement delays, which cannot serve the determination of rock abrasiveness in field. Digital drilling technology is an efficient in situ test method that can effectively link drilling parameters with rock properties. Based on the test results for the CAI and physical and mechanical properties of 13 groups of rock samples collected from southwestern China, this paper focuses on the correlations of the CAI value with rock strength, petrographic characteristics and drilling parameters. By adopting a flat bottom diamond bit that mainly relies on grinding, according to the principle of mechanical balance and energy conservation during rock drilling, the grinding energy per unit volume of rock (ηe) is defined and derived. The univariate regression analysis results show that the uniaxial compressive strength (UCS) and equivalent quartz content (EQC) are the most important factors for explaining CAI and that the CAI can be well estimated using the ηe. Additionally, two optimal regression models for predicting CAI were established using stepwise multiple regression analysis. UCS and ηe were introduced into the model, and the use of both can predict approximately 96% of the variance in the CAI. In addition, a performance evaluation of the models proposed in this paper and previously published CAI prediction models that the prediction models established in this paper are verify statistically more reasonable and reliable than the previously proposed models. The research methods and results provide a new method for the rapid and accurate determination of rock abrasiveness in engineering field.