A New Wear Prediction Method for Degrading Linear Motion Ball Guides Based on Universal Kriging

Abstract Wear-induced degradation in linear motion ball guides (LMBGs) can significantly impact the precision of computer numerical control machine tools. The wear coefficient is a critical parameter determining the wear-rate. However, limited research has been devoted to its experimental investigation in LMBGs. This study proposes a novel wear prediction method for LMBGs that dynamically updates the wear coefficient based on varying contact conditions. To accurately determine the wear coefficient under specific contact parameters, a new experimental approach is introduced, enabling precise measurement of LMBG wear coefficients. Building on these measurements, universal kriging is employed to establish a continuous mapping between the wear coefficient and contact conditions while also providing an uncertainty estimation. Experimental validation confirms the effectiveness of the proposed wear prediction method, which accounts for the variation of the wear coefficient with contact parameters. The constructed wear coefficient map serves as a valuable reference for selecting appropriate wear coefficients in LMBGs. Moreover, by integrating universal kriging, the proposed wear prediction method enables accurate estimation of the effective useful life interval, providing a reliable reference for optimizing the maintenance strategy of LMBGs.