Prediction of Kerf Width in Laser Cutting of Thin Non-Oriented Electrical Steel Sheets Using Convolutional Neural Network
Kerf width is one of the most important quality items in cutting of thin metallic sheets. The aim of this study was to develop a convolutional neural network (CNN) model for analysis and prediction of kerf width in laser cutting of thin non-oriented electrical steel sheets. Three input process param...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-09-01
|
Series: | Mathematics |
Subjects: | |
Online Access: | https://www.mdpi.com/2227-7390/9/18/2261 |
id |
doaj-cfc3d2067d694063888005311dca86e9 |
---|---|
record_format |
Article |
spelling |
doaj-cfc3d2067d694063888005311dca86e92021-09-26T00:38:18ZengMDPI AGMathematics2227-73902021-09-0192261226110.3390/math9182261Prediction of Kerf Width in Laser Cutting of Thin Non-Oriented Electrical Steel Sheets Using Convolutional Neural NetworkDinh-Tu Nguyen0Jeng-Rong Ho1Pi-Cheng Tung2Chih-Kuang Lin3Department of Mechanical Engineering, National Central University, Jhong-Li District, Tao-Yuan City 32001, TaiwanDepartment of Mechanical Engineering, National Central University, Jhong-Li District, Tao-Yuan City 32001, TaiwanDepartment of Mechanical Engineering, National Central University, Jhong-Li District, Tao-Yuan City 32001, TaiwanDepartment of Mechanical Engineering, National Central University, Jhong-Li District, Tao-Yuan City 32001, TaiwanKerf width is one of the most important quality items in cutting of thin metallic sheets. The aim of this study was to develop a convolutional neural network (CNN) model for analysis and prediction of kerf width in laser cutting of thin non-oriented electrical steel sheets. Three input process parameters were considered, namely, laser power, cutting speed, and pulse frequency, while one output parameter, kerf width, was evaluated. In total, 40 sets of experimental data were obtained for development of the CNN model, including 36 sets for training with <i>k</i>-fold cross-validation and four sets for testing. Compared with a deep neural network (DNN) model and an extreme learning machine (ELM) model, the developed CNN model had the lowest mean absolute percentage error (MAPE) of 4.76% for the final test dataset in predicting kerf width. This indicates that the proposed CNN model is an appropriate model for kerf width prediction in laser cutting of thin non-oriented electrical steel sheets.https://www.mdpi.com/2227-7390/9/18/2261laser cuttingkerf widthconvolutional neural networknon-oriented electrical steel |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Dinh-Tu Nguyen Jeng-Rong Ho Pi-Cheng Tung Chih-Kuang Lin |
spellingShingle |
Dinh-Tu Nguyen Jeng-Rong Ho Pi-Cheng Tung Chih-Kuang Lin Prediction of Kerf Width in Laser Cutting of Thin Non-Oriented Electrical Steel Sheets Using Convolutional Neural Network Mathematics laser cutting kerf width convolutional neural network non-oriented electrical steel |
author_facet |
Dinh-Tu Nguyen Jeng-Rong Ho Pi-Cheng Tung Chih-Kuang Lin |
author_sort |
Dinh-Tu Nguyen |
title |
Prediction of Kerf Width in Laser Cutting of Thin Non-Oriented Electrical Steel Sheets Using Convolutional Neural Network |
title_short |
Prediction of Kerf Width in Laser Cutting of Thin Non-Oriented Electrical Steel Sheets Using Convolutional Neural Network |
title_full |
Prediction of Kerf Width in Laser Cutting of Thin Non-Oriented Electrical Steel Sheets Using Convolutional Neural Network |
title_fullStr |
Prediction of Kerf Width in Laser Cutting of Thin Non-Oriented Electrical Steel Sheets Using Convolutional Neural Network |
title_full_unstemmed |
Prediction of Kerf Width in Laser Cutting of Thin Non-Oriented Electrical Steel Sheets Using Convolutional Neural Network |
title_sort |
prediction of kerf width in laser cutting of thin non-oriented electrical steel sheets using convolutional neural network |
publisher |
MDPI AG |
series |
Mathematics |
issn |
2227-7390 |
publishDate |
2021-09-01 |
description |
Kerf width is one of the most important quality items in cutting of thin metallic sheets. The aim of this study was to develop a convolutional neural network (CNN) model for analysis and prediction of kerf width in laser cutting of thin non-oriented electrical steel sheets. Three input process parameters were considered, namely, laser power, cutting speed, and pulse frequency, while one output parameter, kerf width, was evaluated. In total, 40 sets of experimental data were obtained for development of the CNN model, including 36 sets for training with <i>k</i>-fold cross-validation and four sets for testing. Compared with a deep neural network (DNN) model and an extreme learning machine (ELM) model, the developed CNN model had the lowest mean absolute percentage error (MAPE) of 4.76% for the final test dataset in predicting kerf width. This indicates that the proposed CNN model is an appropriate model for kerf width prediction in laser cutting of thin non-oriented electrical steel sheets. |
topic |
laser cutting kerf width convolutional neural network non-oriented electrical steel |
url |
https://www.mdpi.com/2227-7390/9/18/2261 |
work_keys_str_mv |
AT dinhtunguyen predictionofkerfwidthinlasercuttingofthinnonorientedelectricalsteelsheetsusingconvolutionalneuralnetwork AT jengrongho predictionofkerfwidthinlasercuttingofthinnonorientedelectricalsteelsheetsusingconvolutionalneuralnetwork AT pichengtung predictionofkerfwidthinlasercuttingofthinnonorientedelectricalsteelsheetsusingconvolutionalneuralnetwork AT chihkuanglin predictionofkerfwidthinlasercuttingofthinnonorientedelectricalsteelsheetsusingconvolutionalneuralnetwork |
_version_ |
1716870206005968896 |