Identification of Ultra-Short Laser Parameters for a 3D Model of a Thin Metal Film Using the Lattice Boltzmann Method

The paper uses the model of the coupled Boltzmann transport equations, numerically modeled using the lattice Boltzmann method, which describes the impact of an ultra-short laser on thin metal surfaces. The main reason for using this method is the need to create an appropriate model on a very small scale, both in terms of space and time. In the present study, a three-dimensional model is used, for which, in the case of optimization or identification tasks, the characteristics of the obtained numerical solution are quite different from those of other numerical methods and models. The proposed and numerically implemented task of identifying parameters characterizing the laser pulse, such as the power of the stationary laser source, the laser-beam absorptivity coefficient, and the radius of the laser beam, appropriately illustrates the problem. The problem has also been solved for ideal deterministic (no noise) and randomly disturbed (with noise) values of measured temperatures. Three different optimization algorithms are used to solve the inverse task. Several variants of the identification tasks, differing in terms of the number of temperature measurement points, are solved. The results and effectiveness of the identification tasks are compared for the best solution, as well as the statistical measures.