Controllable fabrication of microchannels on silica glass by femtosecond laser: From single-objective to multi-objective collaborative optimization

In ultra-precision machining, the rapid production of high-quality components is essential for enhancing competitiveness. This study applies the Taguchi method and grey relational analysis (GRA) to optimize femtosecond laser processing of microchannels on silica glass. The Taguchi method, widely used in quality engineering and optimization, systematically designs experiments using orthogonal arrays and signal-to-noise ratio analysis is then applied to determine the parameter combination that minimizes surface roughness (S_a) and maximizes material removal rate (MRR). Since S_a and MRR often present conflicting objectives, GRA is employed to address this multi-objective optimization challenge. By adjusting the weight coefficients for quality and efficiency, GRA enables the identification of optimal processing parameters. Under equal weight conditions, the optimal parameters were found to be a pulse energy of 70 μJ, a scanning speed of 774 mm/s, 50 passes, and a scanning interval of 14.1 μm, resulting in an S_a of 2.674 μm and an MRR of 0.0304 mm³/s. The experimental results further validate the reliability of GRA under different weight coefficients This research provides valuable insights into the future advancements in ultra-precision machining technologies.