A Combined CFD, ANFIS and NSGA-II Model for Repetitive Pulse Laser Drilling Process

The article is focused on maximizing the dimensions of a hole drilled through repetitive pulsed laser drilling process by employing the artificial neuro-fuzzy inference system (ANFIS) and multi-objective Genetic Algorithm (GA) with the help of the data obtained from numerical modelling. An axisymmetric two-dimensional numerical model is developed for studying this repetitive pulse laser drilling process in a cylindrical work piece to determine the drilled hole depth and radius and the temperature at the point of laser application. The governing differential equations are discretized using the Finite volume method and the Tri-Diagonal Matrix Algorithm is used to solve the resulting algebraic equations to obtain temperature distribution inside the computational domain. The Enthalpy-Porosity method is used to track the solid-liquid interface during the melting process. The laser source considered here is Gaussian and volumetric in nature and the computations are carried out taking different laser parameters like energy, number of pulses, pulse width, duty cycle and beam radius. The artificial intelligence method ANFIS is adopted to model the process using the result data obtained from numerical investigation. Separate ANFIS models are developed for the prediction of the depth, radius, center temperature, HAZ and taperness of the simulated drilled hole. These models are used as objective functions for a multi-objective genetic algorithm program to get a set optimal solutions for maximizing the depth and radius of the drilled hole and the temperature at the point of laser application in an objective to achieve highest material removal. Another section is also presented carrying out the multi-objective optimisation of considering maximiisation of radius, depth and center temperature along with minimisation of the HAZ and the tapoerness of the hole.