Research on Identification of Key Geometric Errors in Five-Axis Machine Tools Based on Global Sensitivity Analysis

The geometric error (GE) accounts for a significant factor affecting the machine tool’s machining accuracy, and in most cases, large GEs will result in a substantial deviation from the required shape of the machined workpiece. GEs are often observed in five-axis machine tools, and identifying and measuring these errors turns out to be challenging. In the present work, we proposed a novel GE identification approach based on simulations and tests conducted on a BC-type dual-rotary five-axis machine tool. Specifically, a machine tool volumetric error model (VEM), incorporating 41 GEs (the complete model), was constructed using the homogeneous coordinate transformation approach. Then, Sobol sensitivity analysis in conjunction with quasi-Monte Carlo estimation was introduced to the VEM to measure how much each GE contributed to the total volumetric error. The subsequent analysis identified 21 key geometric errors (KGEs). We also compared the simplified VEM and the complete model, and it was revealed that there was little difference between the two, which confirmed the effectiveness of our method. The present work is intended to provide a reference for simplifying VEMs, error element identification, and error compensation.