Research on material removal process and surface morphology evolution of polycrystalline diamond grinding based on multi-sensor fusion

Due to its extreme hardness and brittleness, polycrystalline diamond (PCD) is highly susceptible to surface damage during the grinding process. However, the online monitoring and dynamic evolution process of its microscopic material removal mechanisms have not yet been fully revealed. This paper proposes a grinding state monitoring and morphological characterization method for PCD based on the fusion of force and acoustic signals. Firstly, a theoretical model of the grinding force is constructed. Comparisons between experiments and simulations indicate that the model can accurately predict the evolution laws wherein the grinding force increases nonlinearly with the depth of cut and decreases with the spindle speed. Regarding acoustic emission (AE) signal processing, the effects of grinding parameters on the time and frequency domains of the AE signals are analyzed. Through an in-depth analysis of the power spectrum and energy distribution evolution, the energy dissipation laws at different machining stages are revealed. Meanwhile, the Variational Mode Decomposition (VMD) algorithm is introduced for refined frequency-domain decoupling, adaptively extracting the core frequency band that characterizes the material removal mechanisms. Furthermore, by deeply cross-validating the data-driven AF-RA (Average Frequency - Rise Time to Amplitude ratio) clustering analysis with SEM 3D morphological observations, the optimal linear classification boundary for distinguishing different microscopic damage mechanisms is established. The results demonstrate that as grinding proceeds, the material removal smoothly transitions from brittle fracture dominance in the initial stage to smooth removal dominance in the later stage, resulting in an improved surface roughness. This study reveals the mapping laws among “process parameters - force/acoustic signals - surface morphology.” The proposed method effectively monitors the evolutionary forms of PCD during the ultra-precision grinding process, simultaneously providing multi-dimensional experimental support for the fundamental investigation of PCD grinding mechanisms.