During robotic weld grinding, the complex stress and velocity distribution in the contact zone between the grinding disc and the workpiece directly governs the material removal depth (MRD), which ultimately determines the weld seam profile accuracy and surface quality. Therefore, investigating the material removal mechanism in weld grinding is of great significance for achieving precise process control. Firstly, this study establishes a computational model for the MRD and contour in the semi-elliptical contact area of end-face grinding, based on the inclined contact geometry between the disc and a planar surface, and integrates Preston’s wear theory with Hertzian contact mechanics. Furthermore, the influence of grinding process parameters on the pressure distribution within the contact zone is elucidated. Considering the characteristics of robotic weld grinding, a relative sliding velocity distribution model applicable to both linear and curved paths is derived. Building on these findings, the accumulated MRD under the action of a moving distributed source in the semi-elliptical contact zone is solved via parametric integration. Finally, the results show that under the specified grinding parameters, the predicted MRD from the model and the experimentally measured MRD exhibit consistent variation trends and reach their extreme values at corresponding locations. Across three representative working conditions, the relative error between the predicted and measured maximum removal depth remains below 10%, with the root mean square error not exceeding 0.018 mm. A comparative analysis of the maximum removal depth under different normal forces demonstrates good agreement in both magnitude and trend between theoretical predictions and experimental results. The correctness and applicability of the proposed theoretical model are further verified through comparison with existing distributed contact models and Hertz contact models. This study proposes a novel method for calculating the MRD in the actual semi-elliptical grinding zone considering disc inclination, which holds important theoretical significance for achieving quantitative and precise control in robotic grinding processes.
Yang et al. (Sun,) studied this question.