Machine Learning-Based Prediction of Ablation Groove Geometry and Heat-Affected Zone Formation in Femtosecond Laser Micromachining of Aluminum.

This study presents an Artificial Neural Network (ANN) approach for predicting laser-induced material modifications during femtosecond laser micromachining of aluminum. Experimental investigations were carried out to determine the influence of the average laser power and scanning speed on the width of the ablation groove and the size of the optically determined surface-discoloration width used as a proxy for the Heat-Affected Zone (HAZ). The collected dataset, consisting of 100 samples, was used to develop, train, validate, and test an ANN predictive model with two inputs, two outputs, and two hidden layers. Despite its simplicity and the relatively small dataset, the developed model achieved relatively good prediction accuracy, with an overall correlation coefficient (R) of approximately 0.95 on the test dataset. The predicted values showed reasonable agreement with the experimental results, indicating that the ANN approximated the relationship between laser processing parameters and the resulting material modifications. The presented methodology may provide a useful tool for predicting surface morphology changes and thermal effects in femtosecond laser processing of aluminum.