Optimization of Magnetic Abrasive Finishing Parameters for Co–Cr Alloy Vascular Stent Tubing Using PSO-SVM

To achieve accurate prediction of surface roughness (Ra) in magnetic abrasive finishing (MAF) of the inner wall of Co–Cr alloy vascular stent tubing, and to obtain the optimal process parameter combination for improving the inner surface quality, iron-based diamond magnetic abrasive powders (MAPs) were prepared via plasma melting, centrifugal spraying and rapid solidification. MAF experiments were conducted on Co–Cr alloy vascular stent tubing with an inner diameter of 1.6 mm and an outer diameter of 1.8 mm, and the effects of tube rotational speed, magnetic pole feed rate, abrasive particle size and working gap on surface roughness were investigated. An orthogonal experiment was designed, and a surface roughness prediction model based on particle swarm optimization (PSO) and support vector machine (SVM) was established. Simulation results indicate that the proposed PSO-SVM surface roughness prediction model achieved a coefficient of determination (R2) of 0.96771, a root-mean-square error (RMSE) of 0.0012756 μm, and a mean absolute percentage error (MAPE) of 1.061%. The optimal parameter combination obtained by PSO-SVM optimization was a tube rotational speed of 832.6384 r·min−1, a magnetic pole feed rate of 129.6784 mm·min−1, a working gap of 0.5324 mm, and an abrasive particle size of 132.4185 µm. Under these conditions, the experimentally obtained surface roughness was 0.0949 μm, with a relative error of 0.58% compared to the model-predicted value. The results demonstrate that the established PSO-SVM surface roughness prediction model possesses favorable predictive capability, and its combination with MAF technology enables superior surface quality.