Automatic fault detection for laser powder-bed fusion using semi-supervised machine learning

Abstract Risk-averse areas such as the medical, aerospace and energy sectors have been somewhat slow towards accepting and applying Additive Manufacturing (AM) in many of their value chains. This is partly because there are still significant uncertainties concerning the quality of AM builds. This paper introduces a machine learning algorithm for the automatic detection of faults in AM products. The approach is semi-supervised in that, during training, it is able to use data from both builds where the resulting components were certified and builds where the quality of the resulting components is unknown. This makes the approach cost efficient, particularly in scenarios where part certification is costly and time consuming. The study specifically analyses Laser Powder-Bed Fusion (L-PBF) builds. Key features are extracted from large sets of photodiode data, obtained during the building of 49 tensile test bars. Ultimate tensile strength (UTS) tests were then used to categorise each bar as ‘faulty’ or ‘acceptable’. Using a variety of approaches (Receiver Operating Characteristic (ROC) curves and 2-fold cross-validation), it is shown that, despite utilising a fraction of the available certification data, the semi-supervised approach can achieve results comparable to a benchmark case where all data points are labelled. The results show that semi-supervised learning is a promising approach for the automatic certification of AM builds that can be implemented at a fraction of the cost currently required.

[1]  Dimitris Achlioptas,et al.  Database-friendly random projections: Johnson-Lindenstrauss with binary coins , 2003, J. Comput. Syst. Sci..

[2]  Han Tong Loh,et al.  A Six-sigma approach for benchmarking of RP&M processes , 2006 .

[3]  Jean-Pierre Kruth,et al.  In situ quality control of the selective laser melting process using a high-speed, real-time melt pool monitoring system , 2014 .

[4]  B. Colosimo,et al.  Process defects and in situ monitoring methods in metal powder bed fusion: a review , 2017 .

[5]  Gert Goch,et al.  Development of an adaptive, self-learning control concept for an additive manufacturing process , 2017 .

[6]  Geok Soon Hong,et al.  Defect detection in selective laser melting technology by acoustic signals with deep belief networks , 2018, The International Journal of Advanced Manufacturing Technology.

[7]  Jian Yu,et al.  On convergence and parameter selection of the EM and DA-EM algorithms for Gaussian mixtures , 2018, Pattern Recognit..

[8]  Mohammad Marufuzzaman,et al.  In-situ monitoring of melt pool images for porosity prediction in directed energy deposition processes , 2019, IISE Trans..

[9]  Alaa Elwany,et al.  Prediction of porosity in metal-based additive manufacturing using spatial Gaussian process models , 2016 .

[10]  Amit Mitra,et al.  Statistical Quality Control , 2002, Technometrics.

[11]  Ronay Ak,et al.  A Survey of the Advancing Use and Development of Machine Learning in Smart Manufacturing. , 2018, Journal of manufacturing systems.

[12]  Ge Yu,et al.  Semi-supervised multi-graph classification using optimal feature selection and extreme learning machine , 2018, Neurocomputing.

[13]  Yuhui Zheng,et al.  Automatic segmentation for brain MR images via a convex optimized segmentation and bias field correction coupled model. , 2014, Magnetic resonance imaging.

[14]  Zhiqiang Ge,et al.  Review on data-driven modeling and monitoring for plant-wide industrial processes , 2017 .

[15]  Giuseppe Casalino,et al.  [INVITED] Computational intelligence for smart laser materials processing , 2018 .

[16]  J. Kruth,et al.  Feedback control of selective laser melting , 2007 .

[17]  S. Gold,et al.  In-process sensing in selective laser melting (SLM) additive manufacturing , 2016, Integrating Materials and Manufacturing Innovation.

[18]  Yongsheng Ding,et al.  Multi-model multivariate Gaussian process modelling with correlated noises , 2017 .

[19]  Sushant Negi,et al.  Determining the effect of sintering conditions on mechanical properties of laser sintered glass filled polyamide parts using RSM , 2015 .

[20]  Milton Pereira,et al.  Process observation in fiber laser–based selective laser melting , 2014 .

[21]  Kenneth W. Dalgarno,et al.  An overview of powder granulometry on feedstock and part performance in the selective laser melting process , 2017 .

[22]  Piotr Indyk,et al.  Approximate nearest neighbors: towards removing the curse of dimensionality , 1998, STOC '98.

[23]  Fengqi You,et al.  Data-driven decision making under uncertainty integrating robust optimization with principal component analysis and kernel smoothing methods , 2018, Comput. Chem. Eng..

[24]  Feng Ding,et al.  Expectation maximization estimation algorithm for Hammerstein models with non-Gaussian noise and random time delay from dual-rate sampled-data , 2018, Digit. Signal Process..

[25]  S. Atre,et al.  Effect of process parameters on the Selective Laser Melting (SLM) of tungsten , 2018 .

[26]  M. Zeng,et al.  Wear behavior of Al-Sn alloys with different distribution of Sn dispersoids manipulated by mechanical alloying and sintering , 2008 .

[27]  Nick Jones,et al.  Metal Based Additive Layer Manufacturing: Variations, Correlations and Process Control , 2016, KES.

[28]  B. Colosimo,et al.  Characterization of in-situ measurements based on layerwise imaging in laser powder bed fusion , 2018, Additive Manufacturing.

[29]  S. Shevchik,et al.  Acoustic emission for in situ quality monitoring in additive manufacturing using spectral convolutional neural networks , 2017 .

[30]  James M. Fraser,et al.  In situ morphology-based defect detection of selective laser melting through inline coherent imaging , 2016 .

[31]  Ratna Babu Chinnam,et al.  Observational data-driven modeling and optimization of manufacturing processes , 2017, Expert Syst. Appl..

[32]  Katharina Morik,et al.  Quality Prediction in Interlinked Manufacturing Processes based on Supervised & Unsupervised Machine Learning , 2013 .

[33]  Quirico Semeraro,et al.  In-process Monitoring of Selective Laser Melting: Spatial Detection of Defects via Image Data Analysis , 2017 .

[34]  Sheng Liu,et al.  In Situ 3D Monitoring of Geometric Signatures in the Powder-Bed-Fusion Additive Manufacturing Process via Vision Sensing Methods , 2018, Sensors.

[35]  Eckart Uhlmann,et al.  Intelligent Pattern Recognition of a SLM Machine Process and Sensor Data , 2017 .

[36]  S. Pannala,et al.  The metallurgy and processing science of metal additive manufacturing , 2016 .

[37]  M. Marufuzzaman,et al.  Porosity prediction: Supervised-learning of thermal history for direct laser deposition , 2018 .

[38]  Yuhui Zheng,et al.  A robust modified Gaussian mixture model with rough set for image segmentation , 2017, Neurocomputing.

[39]  Dazhong Wu,et al.  Deep learning for smart manufacturing: Methods and applications , 2018, Journal of Manufacturing Systems.

[40]  J. S. Zuback,et al.  Additive manufacturing of metallic components – Process, structure and properties , 2018 .

[41]  Xinhua Wu,et al.  Influences of processing parameters on surface roughness of Hastelloy X produced by selective laser melting , 2017 .

[42]  Chunbo Zhang,et al.  A Coupled Thermal-Mechanical Analysis of Ultrasonic Bonding Mechanism , 2009 .

[43]  Hai Jin,et al.  MsLRR: A Unified Multiscale Low-Rank Representation for Image Segmentation , 2014, IEEE Transactions on Image Processing.

[44]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[45]  Jean-Pierre Kruth,et al.  Feedback control of Layerwise Laser Melting using optical sensors , 2010 .

[46]  Shashi Phoha,et al.  Application of supervised machine learning for defect detection during metallic powder bed fusion additive manufacturing using high resolution imaging. , 2018 .

[47]  Nathan Halko,et al.  An Algorithm for the Principal Component Analysis of Large Data Sets , 2010, SIAM J. Sci. Comput..

[48]  J. Ziegert,et al.  In situ surface topography of laser powder bed fusion using fringe projection , 2016 .

[49]  Bernardete Ribeiro,et al.  Support vector machines for quality monitoring in a plastic injection molding process , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[50]  G. McLachlan,et al.  The EM algorithm and extensions , 1996 .

[51]  Alireza Karimi,et al.  A data-driven approach to robust control of multivariable systems by convex optimization , 2016, Autom..

[52]  E. O. Olakanmi,et al.  A review on selective laser sintering/melting (SLS/SLM) of aluminium alloy powders: Processing, microstructure, and properties , 2015 .

[53]  Han Tong Loh,et al.  An approach to minimize build errors in direct metal laser sintering , 2006, IEEE Transactions on Automation Science and Engineering.

[54]  I. Yadroitsava,et al.  Selective laser melting of Ti6Al4V alloy for biomedical applications: Temperature monitoring and microstructural evolution , 2014 .

[55]  Fei Tao,et al.  Digital twin-driven product design, manufacturing and service with big data , 2017, The International Journal of Advanced Manufacturing Technology.

[56]  Jack Beuth,et al.  Anomaly Detection and Classification in a Laser Powder Bed Additive Manufacturing Process using a Trained Computer Vision Algorithm , 2018 .

[57]  Syed H. Masood,et al.  Effect of scan strategy on density and metallurgical properties of 17-4PH parts printed by Selective Laser Melting (SLM) , 2017 .