A robust and effective roughness prediction model for laser polishing
摘要
Laser polishing has increasingly emerged as a promising technique owing to its unique advantages, such as high flexibility and strong automation potential. However, modelling and predicting its outcomes remain challenging due to the complex physical phenomena involved, including free surface flow and multiple phase interfaces. In this context, Smooth Particle Hydrodynamics (SPH), a mesh-free numerical method, offers strong potential for simulating such processes. The present study aims to explore this potential by developing an effective SPH model for predicting surface roughness. In the developed model, the key thermal and fluid dynamic behaviours during the laser polishing process are considered. Furthermore, a geometry-based method for curvature estimation is implemented to overcome the limitations of conventional SPH approaches. To ensure accuracy, the model is first numerically benchmarked, before being fully implemented for the laser polishing of DIN 1.2379 tool steel. In the laser polishing simulations, the melt depth and the transient evolution of the surface profile are analysed. Upon complete solidification of the polished workpieces, the simulated average surface roughness (Ra) values are validated against experimental measurements reported in the literature. The simulated Ra values show good agreement with the experimental data at laser powers ranging from 400 W to 750 W. In particular, the predicted Ra reduction presents an average deviation of only 8% from the experimental measurements.