A study on the wear process of impregnated diamond bits via acoustic emission

Abstract

Impregnated diamond (ID) drilling is a successful and widely used technique for drilling abra-sive and hard rock formations. This is in part due to their self-sharpening mechanism, which consists of a con-stant renewal of diamonds acting in the cutting face as matrix wear takes place. This mechanism, however, can be altered by the blunting of the bit. Therefore, our objective in this paper is to evaluate the feasibility of artificial intelligence (AI) based techniques to monitor tool condition of ID bits, sharp or blunt. Accordingly, topologically-invariant tests are performed with sharp and blunt bits while recording acoustic emissions (AE) and measuring-while-drilling variables. These are then utilised as inputs variables to the tool condition moni-toring system. Acceptable pattern recognition rates were obtained for multiple input variables combinations with different pattern recognition algorithms, particularly, that composed by AE root mean square (AErms) and torque-on-bit (tob).