Laser Ablation of Cobalt-Bound Tungsten Carbide and Aluminium Oxide Ceramic: Experimental Investigation with ANN Modelling and GA Optimisation
This paper reports the results of an experimental investigation into the ablation of cobalt-bound tungsten carbide and aluminium oxide ceramic, two so-called super-composite materials, using a nanosecond pulsed laser. The ablation of single trenches was performed for different scan speeds and laser fluence. The ablation process was assessed in terms of the surface roughness along the centre lines of the trenches and the ablation rate. Using an Artificial Neural Network (ANN) and Genetic Algorithm (GA), a model was generated using the laser parameters as inputs and measured results as outputs. The optimal results predicted by the model were validated experimentally with a maximum difference of 14% between predicted values and measured results. The model correctly quantified the effect of the laser settings (fluence and scan speed) on surface roughness and ablation rate and identified the processing window and ablation conditions, for the optimum ablation performance of a nanosecond pulsed laser. Nevertheless, with the surface roughness of the aluminium oxide ceramic there was a noticeable difference of 32.7% between the prediction by the model and the results of the validation test, an exception to generally accurate predictions. Modelling of the laser ablation process reduces the number of trials during the setup stage, saving time and contributing to a more efficient and economically sustainable manufacturing process.
Keywordsns pulsed laser ablation Ablation rate Surface roughness ANN GA
This research was funded by the Steep project “A Synergetic Training Network on Energy Beam Processing: from modelling to industrial applications” within EC seventh framework program under the grant agreement no 316560.
The authors gratefully acknowledge the technical support of Dr Eva Rodriguez and Mr Jon Etxarri from IK4-Tekinker, Spain.
- 5.Elkaseer, A., Lambarri, J., Sarasua, J., Cascón, I.: On the development of a chip breaker in a metal-matrix polycrystalline diamond insert: finite element based design with ns-laser ablation and machining verification. J. Micro Nano-Manuf. 5(3), 031007–031007-12 (2017). https://doi.org/10.1115/1.4036933
- 12.Burden, F., Winkler, D.: Bayesian regularization of neural networks. In: Livingstone, D.J. (ed.) Artificial Neural Networks. Methods in Molecular Biology™, vol. 458. Humana Press, New York (2008)Google Scholar
- 13.MathWorks website. https://www.mathworks.com/help/nnet/ref/trainbr.html. Accessed 17 Apr 2018
- 14.MathWorks website. https://www.mathworks.com/help/gads/what-is-the-genetic-algorithm.html. Accessed 17 Apr 2018