Silicone-replication-assisted surface topography characterization of fine-grained diamond grinding wheels using density-based clustering

Fine-grained diamond grinding wheels have been widely employed in the precision machining of hard and brittle materials owing to their superior grinding performance. The distribution parameters of abrasive grains, such as protrusion height and grain packing density, play a decisive role in grinding quality and efficiency. However, due to the large size and complex surface topographies, it is difficult for conventional measurement methods to achieve high-accuracy and nondestructive characterization of grinding wheel topography. To address this challenge, this study proposed a surface topography characterization method for fine-grained diamond grinding wheels based on silicone replication and density-based clustering. The surface topography of the wheel was replicated using a high-precision silicone molding technique, and the replica surface was measured by a laser confocal microscope. Curvature correction, burr removal, and an adaptive filtering method based on power spectral density (PSD) were then applied to enhance data quality. Abrasive grains were accurately identified by the density-based spatial clustering of applications with noise (DBSCAN) DBSCAN algorithm, enabling the statistical extraction of grain-distribution parameters. Experimental results demonstrated that the replication accuracy evaluated by surface roughness showed low relative errors (4.8% for Ra and 3.7% for Rq). The proposed method effectively identified the grain protrusion height of a D7 diamond grinding wheel, with an average value of approximately 20% ∼ 30% of the nominal grain size. Furthermore, the effects of mechanical dressing parameters on abrasive protrusion height were investigated. Increasing the compensation coefficient, selecting a lower dressing wheel hardness, and employing an appropriate dressing wheel grain size yielded larger protrusion heights and lower surface roughness of workpiece.