Presentation Information
[23p-13M-15]Machine Learning-Based Prediction of Laser-Processing Parameters for Fabricating Optical Diffusers
〇(B)Ryoma Kawaoto1, Tomotaro Namba1, Yukiyoshi Ohtsuki1, Feng Yan2, Takashi Nakajima2 (1.Tohoku Univ., 2.Kyoto Univ.)
Keywords:
machine learning,laser processing,optical diffuser
For simple and cost-effective fabrication of glass optical diffusers, the combined dry process which consists of ablation-assisted nanosecond laser micromachining and laser cleaning was developed. However, the process involves the complicated procedures to find suitable sets of fabrication parameters. To overcome this difficulty, we adopt a machine-learning approach for the sake of the fabrication parameter screening. As a first step, we develop the machine-learning approach to predict transmitted beam profiles through the optical diffusers from fabrication parameters.