A Layer-Wise Surface Deformation Defect Detection by Convolutional Neural Networks in Laser Powder-Bed Fusion Images
Surface deformation is a multi-factor, laser powder-bed fusion (LPBF) defect that cannot be avoided entirely using current monitoring systems. Distortion and warping, if left unchecked, can compromise the mechanical and physical properties resulting in a build with an undesired geometry. Increasing...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-10-01
|
Series: | Materials |
Subjects: | |
Online Access: | https://www.mdpi.com/1996-1944/15/20/7166 |
_version_ | 1797471771559985152 |
---|---|
author | Muhammad Ayub Ansari Andrew Crampton Simon Parkinson |
author_facet | Muhammad Ayub Ansari Andrew Crampton Simon Parkinson |
author_sort | Muhammad Ayub Ansari |
collection | DOAJ |
description | Surface deformation is a multi-factor, laser powder-bed fusion (LPBF) defect that cannot be avoided entirely using current monitoring systems. Distortion and warping, if left unchecked, can compromise the mechanical and physical properties resulting in a build with an undesired geometry. Increasing dwell time, pre-heating the substrate, and selecting appropriate values for the printing parameters are common ways to combat surface deformation. However, the absence of real-time detection and correction of surface deformation is a crucial LPBF problem. In this work, we propose a novel approach to identifying surface deformation problems from powder-bed images in real time by employing a convolutional neural network-based solution. Identifying surface deformation from powder-bed images is a significant step toward real-time monitoring of LPBF. Thirteen bars, with overhangs, were printed to simulate surface deformation defects naturally. The carefully chosen geometric design overcomes problems relating to unlabelled data by providing both normal and defective examples for the model to train. To improve the quality and robustness of the model, we employed several deep learning techniques such as data augmentation and various model evaluation criteria. Our model is 99% accurate in identifying the surface distortion from powder-bed images. |
first_indexed | 2024-03-09T19:53:46Z |
format | Article |
id | doaj.art-13ca44399d5d4edcb17122d1200afe7e |
institution | Directory Open Access Journal |
issn | 1996-1944 |
language | English |
last_indexed | 2024-03-09T19:53:46Z |
publishDate | 2022-10-01 |
publisher | MDPI AG |
record_format | Article |
series | Materials |
spelling | doaj.art-13ca44399d5d4edcb17122d1200afe7e2023-11-24T01:03:17ZengMDPI AGMaterials1996-19442022-10-011520716610.3390/ma15207166A Layer-Wise Surface Deformation Defect Detection by Convolutional Neural Networks in Laser Powder-Bed Fusion ImagesMuhammad Ayub Ansari0Andrew Crampton1Simon Parkinson2School of Computing and Engineering, University of Huddersfield, Huddersfield HD1 3DH, UKSchool of Computing and Engineering, University of Huddersfield, Huddersfield HD1 3DH, UKSchool of Computing and Engineering, University of Huddersfield, Huddersfield HD1 3DH, UKSurface deformation is a multi-factor, laser powder-bed fusion (LPBF) defect that cannot be avoided entirely using current monitoring systems. Distortion and warping, if left unchecked, can compromise the mechanical and physical properties resulting in a build with an undesired geometry. Increasing dwell time, pre-heating the substrate, and selecting appropriate values for the printing parameters are common ways to combat surface deformation. However, the absence of real-time detection and correction of surface deformation is a crucial LPBF problem. In this work, we propose a novel approach to identifying surface deformation problems from powder-bed images in real time by employing a convolutional neural network-based solution. Identifying surface deformation from powder-bed images is a significant step toward real-time monitoring of LPBF. Thirteen bars, with overhangs, were printed to simulate surface deformation defects naturally. The carefully chosen geometric design overcomes problems relating to unlabelled data by providing both normal and defective examples for the model to train. To improve the quality and robustness of the model, we employed several deep learning techniques such as data augmentation and various model evaluation criteria. Our model is 99% accurate in identifying the surface distortion from powder-bed images.https://www.mdpi.com/1996-1944/15/20/7166surface deformationLPBFmetal additive manufacturingconvolutional neural networkmachine learningdeep learning |
spellingShingle | Muhammad Ayub Ansari Andrew Crampton Simon Parkinson A Layer-Wise Surface Deformation Defect Detection by Convolutional Neural Networks in Laser Powder-Bed Fusion Images Materials surface deformation LPBF metal additive manufacturing convolutional neural network machine learning deep learning |
title | A Layer-Wise Surface Deformation Defect Detection by Convolutional Neural Networks in Laser Powder-Bed Fusion Images |
title_full | A Layer-Wise Surface Deformation Defect Detection by Convolutional Neural Networks in Laser Powder-Bed Fusion Images |
title_fullStr | A Layer-Wise Surface Deformation Defect Detection by Convolutional Neural Networks in Laser Powder-Bed Fusion Images |
title_full_unstemmed | A Layer-Wise Surface Deformation Defect Detection by Convolutional Neural Networks in Laser Powder-Bed Fusion Images |
title_short | A Layer-Wise Surface Deformation Defect Detection by Convolutional Neural Networks in Laser Powder-Bed Fusion Images |
title_sort | layer wise surface deformation defect detection by convolutional neural networks in laser powder bed fusion images |
topic | surface deformation LPBF metal additive manufacturing convolutional neural network machine learning deep learning |
url | https://www.mdpi.com/1996-1944/15/20/7166 |
work_keys_str_mv | AT muhammadayubansari alayerwisesurfacedeformationdefectdetectionbyconvolutionalneuralnetworksinlaserpowderbedfusionimages AT andrewcrampton alayerwisesurfacedeformationdefectdetectionbyconvolutionalneuralnetworksinlaserpowderbedfusionimages AT simonparkinson alayerwisesurfacedeformationdefectdetectionbyconvolutionalneuralnetworksinlaserpowderbedfusionimages AT muhammadayubansari layerwisesurfacedeformationdefectdetectionbyconvolutionalneuralnetworksinlaserpowderbedfusionimages AT andrewcrampton layerwisesurfacedeformationdefectdetectionbyconvolutionalneuralnetworksinlaserpowderbedfusionimages AT simonparkinson layerwisesurfacedeformationdefectdetectionbyconvolutionalneuralnetworksinlaserpowderbedfusionimages |