Regulating LIPSS Period and Orientation via Multiaxis Laser Processing and Neural Network Guidance

While femtosecond laser-induced periodic surface structures (LIPSS) offer substantial potential in surface engineering, realizing a full spectrum of their demand-oriented functionalities depends critically on the precise and controllable tailoring of their key parameters, period, and orientation. This study proposes a strategy for precisely and simultaneously controlling both LIPSS characteristics by coupling the polarization and incidence angles of the laser beam, which is realized by a multiaxis femtosecond laser micromachining process integrated with neural network guidance. A multiaxis femtosecond laser micromachining system for LIPSS-based patterning is constructed, and its kinematic model is established to compensate for the polarization deviations induced upon the variation of the beam incidence angle. Back-propagation (BP) neural network and particle swarm optimization method is employed to optimize laser processing parameters (pulse energy, effective pulse number, and scanning line spacing) to ensure the LIPSS regularity and uniformity. Another BP neural network is established to elucidate the mapping relationship between laser processing parameters (incidence/polarization angles) and LIPSS geometry, which is experimentally validated by demonstrating the designed multilevel LIPSS-based structural coloring of SUS 304 stainless steel, where the period and orientation are simultaneously tailored. This work provides both a theoretical basis and technical guidance for the design and fabrication of LIPSS with on-demand characteristics.