Comparative Statistical Analysis For Prediction Of Depth Of Cut In Co2 Laser Machining
The present study investigates the relationships and parameter interactions between three controllable variables on depth of cut in CO2 laser cutting of acrylic material using three statistical methods. Laser power, traverse speed and number of pass of laser probe are selected as independent variables. The three statistical methods considered were Full Factorial (FF), Response Surface Method (RSM) and Taguchi method for predicting the relation between response and the variables. The predicted models were developed using adjusted sum of square method and finalized by linear multiple regression technique. The ANOVA test has also been carried out to study the effect of main variables and their interaction effects on the response as well as to check the adequacy of the models at 95% confidence level. The final models were checked by complete residual analysis and finally were experimentally verified. It was found and the average deviation of depth of cut between predicted and actual values with full factorial method was found to be 0.0225 mm, with RSM method, it was found to be 0.052 mm and with Taguchi, it was -0.07667 mm.
Index Terms- Laser beam machining, full factorial method, response surface method, Taguchi method, ANOVA.