Prediction and optimization of dross formation in laser cutting of electrical steel sheet in different environments

Abstract

Dross formation in laser cutting of electrical steel sheet in different environments, namely oil, alcohol, and air, was predicted and optimized using a deep neural network (DNN) and an improved grey wolf optimizer (I-GWO), respectively. Analyses using random forest method and response surface method showed that the cutting environment, laser power, pulse frequency, and cutting speed had a significant influence on the dross formation. In addition, cutting in oil leads to less dross formation than in alcohol and air. A stacked autoencoder method combined with a multi-objective grey wolf optimizer was employed to generate a pre-trained DNN, followed by a fine-tuning process to obtain the final DNN. The I-GWO was used to determine the optimal combination of process parameters for minimum dross formation. The DNN model proved its efficacy by showing very low values of mean absolute percentage error and very high values of absolute fraction of variation for the training, validation, and testing datasets. Moreover, the accuracy of the developed DNN model was higher than that of other artificial intelligence based methods, namely random vector functional link and support vector machine for regression, as evaluated by nine statistical criteria. The predicted optimal process parameters by the DNN and I-GWO algorithms were verified by validation experiments in which the minimum dross formation was generated.