Five-axis machine tools accuracy condition monitoring based on volumetric errors and vector similarity measures

The accuracy of a machine tool affects the geometry and dimensions of machined parts. A machine tool accuracy condition monitoring scheme using volumetric errors (VEs), vector similarity measures (VSMs) and exponentially weighted moving average (EWMA) control chart is proposed in this research. The usefulness of this scheme is tested with simulated machine error data as well as real machine tool tests using NC induced geometric error changes and a real C-axis encoder fault. Both sudden and gradual changes were considered for the simulated faults. The results show that VE is a meaningful quantity for the monitoring of the machine tool accuracy condition. The proposed VSMs work well in VEs feature extraction. Amongst the studied VSMs, the module of the vectorial difference of two consecutive VE vectors (Dist) and the angle between those vectors (Cost) are more stable and perform better for monitoring faults with sudden and gradual changes than the remaining VSMs in real VE data processing. Finally, this research provides guidelines for the use of VEs as well as a VE-based monitoring strategy for monitoring machine tool accuracy condition.