A generalised approach on kerf geometry prediction during CO2 laser cut of PMMA thin plates using neural networks [post]

John Dimitrios Kechagias, Konstantinos Ninikas, Panagiotis Stavropoulos, Konstantinos Salonitis
2021 unpublished
This study presents an application of feedforward and backpropagation neural network (FFBP-NN) for predicting the kerf characteristics, i.e. the kerf width in three different distances from the surface (upper, middle and down) and kerf angle during laser cutting of PMMA thin plates. Stand-off distance, cutting speed and beam power are the studied parameters for the case of low power CO2 laser cutting. A three-parameter three-level full factorial array has been used and twenty-seven (33) cuts
more » ... e performed. Subsequently, the kerf width and angle were measured and analysed through ANOM, ANOVA and interaction plots. The statistical analysis highlighted that linear modeling is insufficient for the precise prediction of kerf characteristics. A FFBP-NN was developed, trained, validated and generalised for the accurate prediction of the kerf geometry. The FFBP-NN achieved an R-sq value of 0.98, in contrast to the ANOVA linear models which achieved a value of about 0.90.
doi:10.21203/rs.3.rs-268745/v1 fatcat:5n242w3unra3tihn4ks5wxncbi