Underwater laser micro-milling of fine-grained aluminium and the process modelling by machine learning

Abstract

Nanosecond-pulsed laser ablation is often accompanied by adverse thermal effects such as oxidation, debris recast and burr formation. To reduce these effects, in this paper, the authors present the underwater laser milling process using RSA-905 fine-grained aluminium as the target material for the first time. The results show that channels up to 200 μm in width, 700 μm depth and bottom roughness around 1 µm Ra could be fabricated with reduced thermal effects. By conducting multi- and single-factor experiments, empirical models relating the laser processing parameters to the key dimensions of channels were derived using artificial neural network (ANN) algorithm and polynomial regression (PR), and the models' accuracies were evaluated. Based on the models, the cross-section profile of a channel subject to a given set of processing parameters can be predicted. The process can serve as a pre-treatment technique of mechanical milling such that the tool life will be extended and the profile of a desired feature can be precisely defined.