摘要
Optimizing laser processes is historically challenging,requiring extensive and costly experimentation.To solve this issue,we apply Bayesian optimization for process parameter optimization to laser cutting,welding,and polishing.We demonstrate how readily available Bayesian optimization frameworks enable efficient optimization of laser processes with only modest expert knowledge.Case studies on laser cutting,welding,and polishing highlight its adaptability to real-world manufacturing scenarios.Moreover,the examples emphasize that with suitable cost functions and boundaries an acceptable optimization result can be achieved after a reasonable number of experiments.
基金
support of the projects within the InnovationsCampus Mobilität der Zukunft as well as for the sustainability support of the projects of the Exzellenzinitiative II.The authors would like to thank Precitec Optronik GmbH(Germany)for providing the OCT sensor Chrocodile2.The authors would like to thank Light Conversion(Lithuania)for providing the Carbide CB3-80 laser.The Laser beam source TruDisk8001(DFG object number:625617)was funded by the Deutsche Forschungsgemeinschaft(DFG,German Research Foundation)-INST 41/990-1 FUGG.
