Parameters optimization of laser transmission welding polycarbonate based on LibSVM

Transparent polycarbonate(PC) was selected as the experimental object and mixtures of aluminum powders and carbon powders were used as the absorbent. The support vector machine algorithm was used to perform regression analysis on the experimental data of laser transmission welding polycarbonate to obtain the theoretical optimal welding parameters. Firstly, the surface response method was used to plan the experiment program to obtain the correlation model of welding quality and welding process. Secondly, LibSVM support vector machine was used to optimize welding parameters(laser power, absorbent ratio, welding speed, surface roughness). Finally, model regression was used to predict and experimental verification was conducted. The research results showed that the best welding quality was obtained under condition of37 W laser power, 5 mm/s welding speed, 29% aluminum powder content and 1. 77 μm surface roughness. At the same time, the error between the optimized predicted value and the experimental value was small. It was of guiding significance to reduce welding cost and improve welding quality and welding efficiency.