Modeling and Optimization on Nd: YAG Laser Marking of Zirconia Ceramic using RSM and ANN

Laser marking is one of the well-developed technologies of materials processing. Laser marking is the best and most applied permanent marking method. The present research papers deals with the artificial neural network (ANN) and the response surface methodology (RSM) based mathematical modelling and also an optimization analysis on marking characteristics i.e., mark width, mark depth and mark intensity on zirconia ceramic. The major influencing laser marking process parameter considered are pulse frequency, lamp current, pulse width, scanning speed and air pressure. The experiments have been planned and carried out based on RSM based modelling with 32 runs. ANN modelling is performed and the results are compared. The average percentage of prediction errors of the developed ANN model for mark width, mark depth and mark intensity are 2.52, 2.58 and 2.58 respectively and the overall percentage of prediction error is 2.6. The output of the RSM optimal data is validated through experimentation and ANN predictive model. A good agreement is observed between the resultsbased on ANN predictive model of 82.8 μm, 46.3μm and 0.605 for mark width, mark depth and mark intensity respectively and actual experimental observations.