Topography Modeling of Surface Grinding Based on Random Abrasives and Performance Evaluation

Abstract

Abstract The surface morphology and roughness of a workpiece are crucial parameters in grinding processes. Accurate prediction of these parameters is essential for maintaining the workpiece’s surface integrity. However, the randomness of abrasive grain shapes and workpiece surface formation behaviors poses significant challenges, and accuracy in current physical mechanism-based predictive models is needed. To address this problem, by using the random plane method and accounting for the random morphology and distribution of abrasive grains, this paper proposes a novel method to model CBN grinding wheels and predict workpiece surface roughness. First, a kinematic model of a single abrasive grain is developed to accurately capture the three-dimensional morphology of the grinding wheel. Next, by formulating an elastic deformation and formation model of the workpiece surface based on Hertz theory, the variation in grinding arc length at different grinding depths is revealed. Subsequently, a predictive model for the surface morphology of the workpiece ground by a single abrasive grain is devised. This model integrates the normal distribution model of abrasive grain size and the spatial distribution model of abrasive grain positions, to elucidate how the circumferential and axial distribution of abrasive grains influences workpiece surface formation. Lastly, by integrating the dynamic effective abrasive grain model, a predictive model for the surface morphology and roughness of the grinding wheel is established. To examine the impact of changing the grit size of the grinding wheel and grinding depth on workpiece surface roughness, and to validate the accuracy of the model, experiments are conducted. Results indicate that the predicted three-dimensional morphology of the grinding wheel and workpiece surfaces closely matches the actual grinding wheel and ground workpiece surfaces, with surface roughness prediction deviations as small as 2.3%.

Keywords

• Surface topography prediction
• Grinding
• Grinding wheel model
• Random plane method