Using four-dimensional geometric models for representing dynamic machining process based on parallel processing by GPGPU

Abstract

Calculating the interaction between a machining tool and a workpiece is very important in evaluating cutting process. The estimation of cutter-workpiece engagement based on three-dimensional (3D) geometric models needs huge amount of discrete data for representing dynamic states of the cutting tool and the workpiece. In the paper, authors propose a Spatio-Temporal description approach to represent dynamic machining process using four-dimensional (4D) geometric models. As the core of the approach, T-map data structure is newly introduced and a calculation of cutter-workpiece engagement using T-map is proposed. To accelerate the calculation with T-map, several parallel processing strategies are compared.