Global sensitivity analysis of a CNC machine tool: application of MDRM

Abstract

Understanding the error propagation mechanism of a CNC machine tool is a prerequisite to improve machine tool kinematic accuracy by other techniques. This paper presents system reliability and global sensitivity analyses of a machine tool with the multiplicative dimensional reduction method (M-DRM). Reliability of the machine tool is defined as the probability that output error remains within a specified limit in entire target trajectory of tool cutter. This reliability problem is formulated as a series reliability problem that is solved using probability distribution of the maximal positioning error. The variance-based global sensitivity analysis is preformed to screen significant error components of the system kinematic error model. The proposed method is illustrated by a three-axis vertical milling machine tool for system reliability and global sensitivity analyses. Results are verified by Monte-Carlo simulation method. The procedure would be easily applied for probability-based kinematic analysis of machine tools based on other error models.