A new mathematical model for optimizing laser cutting parameters to improve fabric quality

Abstract

We present a model for optimizing laser-cutting process parameters to improve fabric-cutting quality in the textile industry. We use two methods to predict fabric-cutting quality based on customized laser-cutting parameters, Response Surface Methodology (RSM) and Artificial Neural Network (ANN). RSM had an R-squared (R2) value of 0.952, showing high accuracy. As a result of varying iterations and nodes, the results of the ANN models were different. As a result of 10,000 iterations on an architecture with six nodes and one hidden layer, the ANN model with an R-squared of 0.998 was the best optimization model. The novelty of this study found that ANNs with six nodes and 10,000 iterations optimize the laser cutting process for fabrics more effectively than models with fewer nodes and fewer iterations. RSM and ANNs are effective tools for improving fabric-cutting quality in specific applications, as well as providing theoretical contributions through this research.