Online prediction of diffusion wear on the flank through tool tip temperature in turning using artificial neural networks

Abstract Diffusion in cutting tool materials during machining contributes significantly to the tool wear and has been drawing the attention of several researchers for over a decade. The present work aims at employing a method to investigate tool material diffusion for online prediction of flank wear using artificial neural networks. Experimentation is carried on a lathe under different cutting conditions. Flank wear is measured at various intervals of time off-line. Back propagation neural network fortified with heuristic methods of optimization has been employed in the present work. The cutting conditions and nodal temperature measured by a remote thermocouple are used as inputs and the amount of diffusion wear is obtained as the output. The model is validated by comparing with the experimental results. The proposed methodology, which employs a diffusion parameter for the assessment of tool wear from estimated tool tip temperatures, can be adapted to any combination of tool and work material.