Geometrical quality prediction of machining process by Exechon X-mini PKM through deformation modelling and error compensation

Abstract

Parallel Kinematic Machines (PKMs) offer enhanced motion dynamics and flexibility, bridging the gap between conventional CNC machines and industrial robots. Stiffness, a key determinant of machining accuracy, is often modelled with limited consideration of gravitational effects, leading to reduced predictive accuracy. This paper introduces a novel stiffness modelling approach that integrates a theoretical model without gravity and gravity-based parameter optimisation through experimental analysis. Comprehensive stiffness measurements were conducted to isolate gravitational effects on the machine structure, enabling precise calibration of the theoretical model for accurate stiffness prediction. A six-dimensional stiffness analysis of the X-Mini machine tool using the optimised model demonstrated improved prediction accuracy, reducing errors by 14 %, 21 %, and 8 % in the X, Y and Z directions, respectively. Predicted stiffness and estimated cutting forces were used to compute workspace deformations, which were then compensated by modifying the depth of cut in slot milling. Experimental validation demonstrated the method’s effectiveness, achieving a machined shape error prediction accuracy of 6–9 µm. This approach can be well applied to shape quality prediction of machined parts by robots and machine tools.<br/><br/>