Workpiece hardness classification in stone drilling using cutting forces and servomotor current signals

Kamen je izrazito heterogen i anizotropan materijal karakteriziran svojim mineralnim sastavom, teksturom i strukturom. Varijacije fizikalno-kemijskih svojstava različitih vrsta kamena imaju značajan utjecaj na dinamiku obradnog procesa. Pritom je tvrdoća jedno od najznačajnijih svojstava kamena s direktnim utjecajem na sile rezanja. Varijacije tvrdoće u zoni rezanja često utječu na povećanu dinamiku trošenja reznog alata, što posljedično može utjecati i na lom alata i/ili obratka. Stoga je određivanje tvrdoće obratka za vrijeme obrade kamena i sličnih izrazito nehomogenih materijala istovremeno izazovan zadatak, ali i nužan segment nadzora obradnih sustava i procesa. Posebnost ovog rada je analiza utjecaja signala sila rezanja i jakosti struja pogonskih motora stroja u klasifikaciji tvrdoće kamenih obradaka bušenih novodizajniranim svrdlom s kanalom za ispuhivanje. Primjena navedene vrste svrdla je u laboratorijskim uvjetima pokazala značajno manje sile rezanja, vibracije i intenzitet trošenja rezne oštrice. Postupak klasifikacije tvrdoće proveden je primjenom neuronske mreže s radijalnim baznim funkcijama, a sa svrhom povećanja robusnosti modela za klasifikaciju istrošenosti alata i tvrdoće kamena.Stone is an extremely heterogeneous and anisotropic material characterized by its mineral composition, texture, and structure. Variations in the physio-chemical properties of different types of stone have a significant impact on the dynamics of the machining process. Hardness is one of the most important properties of stone, directly influencing cutting forces. Variations in hardness within the cutting zone often affect increased dynamics of tool wear, consequently potentially influencing tool breakage and/or workpiece damage. Therefore, determining the hardness of the workpiece during the processing of stone and similar highly heterogeneous materials is simultaneously a challenging task and a necessary aspect of monitoring of machining systems and processes. The objective of this study is to analyse the influence of cutting force and servomotor current signals on the classification of the hardness of stone workpieces drilled with a newly designed drill bit featuring an internal cooling channel. The application of this type of drill bit in laboratory conditions demonstrated significantly reduced cutting forces, vibrations, and cutting-edge wear intensity. The hardness classification procedure was carried out using a neural network with a radial basis function, aiming to enhance the robustness of the model for classifying tool wear and stone hardness