Quality Prediction and Classification of Process Parameterization for Multi-Material Jetting by Means of Computer Vision and Machine Learning
Multi-Material Jetting (MMJ) is an additive manufacturing process empowering the printing of ceramics and hard metals with the highest precision. Given great advantages, it also poses challenges in ensuring the repeatability of part quality due to an inherent broader choice of built strategies. The...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Journal of Manufacturing and Materials Processing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-4494/8/1/8 |
| _version_ | 1797297786890223616 |
|---|---|
| author | Armin Reckert Valentin Lang Steven Weingarten Robert Johne Jan-Hendrik Klein Steffen Ihlenfeldt |
| author_facet | Armin Reckert Valentin Lang Steven Weingarten Robert Johne Jan-Hendrik Klein Steffen Ihlenfeldt |
| author_sort | Armin Reckert |
| collection | DOAJ |
| description | Multi-Material Jetting (MMJ) is an additive manufacturing process empowering the printing of ceramics and hard metals with the highest precision. Given great advantages, it also poses challenges in ensuring the repeatability of part quality due to an inherent broader choice of built strategies. The addition of advanced quality assurance methods can therefore benefit the repeatability of part quality for widespread adoption. In particular, quality defects caused by improperly configured droplet overlap parameterizations, despite droplets themselves being well parameterized, constitute a major challenge for stable process control. This publication deals with the automated classification of the adequacy of process parameterization on green parts based on in-line surface measurements and their processing with machine learning methods, in particular the training of convolutional neural networks. To generate the training data, a demo part structure with eight layers was printed with different overlap settings, scanned, and labeled by process engineers. In particular, models with two convolutional layers and a pooling size of (6, 6) appeared to yield the best accuracies. Models trained only with images of the first layer and without the infill edge obtained validation accuracies of 90%. Consequently, an arbitrary section of the first layer is sufficient to deliver a prediction about the quality of the subsequently printed layers. |
| first_indexed | 2024-03-07T22:26:24Z |
| format | Article |
| id | doaj.art-225637f1bf95450e9bf9f391357bf635 |
| institution | Directory Open Access Journal |
| issn | 2504-4494 |
| language | English |
| last_indexed | 2024-03-07T22:26:24Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Manufacturing and Materials Processing |
| spelling | doaj.art-225637f1bf95450e9bf9f391357bf6352024-02-23T15:22:50ZengMDPI AGJournal of Manufacturing and Materials Processing2504-44942024-01-0181810.3390/jmmp8010008Quality Prediction and Classification of Process Parameterization for Multi-Material Jetting by Means of Computer Vision and Machine LearningArmin Reckert0Valentin Lang1Steven Weingarten2Robert Johne3Jan-Hendrik Klein4Steffen Ihlenfeldt5Institute of Mechatronic Engineering, Technische Universität Dresden TUD, 01069 Dresden, GermanyInstitute of Mechatronic Engineering, Technische Universität Dresden TUD, 01069 Dresden, GermanyFraunhofer-Institut für Keramische Technologien und Systeme IKTS, 01277 Dresden, GermanyFraunhofer-Institut für Keramische Technologien und Systeme IKTS, 01277 Dresden, GermanyInstitute of Mechatronic Engineering, Technische Universität Dresden TUD, 01069 Dresden, GermanyInstitute of Mechatronic Engineering, Technische Universität Dresden TUD, 01069 Dresden, GermanyMulti-Material Jetting (MMJ) is an additive manufacturing process empowering the printing of ceramics and hard metals with the highest precision. Given great advantages, it also poses challenges in ensuring the repeatability of part quality due to an inherent broader choice of built strategies. The addition of advanced quality assurance methods can therefore benefit the repeatability of part quality for widespread adoption. In particular, quality defects caused by improperly configured droplet overlap parameterizations, despite droplets themselves being well parameterized, constitute a major challenge for stable process control. This publication deals with the automated classification of the adequacy of process parameterization on green parts based on in-line surface measurements and their processing with machine learning methods, in particular the training of convolutional neural networks. To generate the training data, a demo part structure with eight layers was printed with different overlap settings, scanned, and labeled by process engineers. In particular, models with two convolutional layers and a pooling size of (6, 6) appeared to yield the best accuracies. Models trained only with images of the first layer and without the infill edge obtained validation accuracies of 90%. Consequently, an arbitrary section of the first layer is sufficient to deliver a prediction about the quality of the subsequently printed layers.https://www.mdpi.com/2504-4494/8/1/8additive manufacturingmulti-material jettingprocess monitoringartificial intelligencemachine learningcomputer vision |
| spellingShingle | Armin Reckert Valentin Lang Steven Weingarten Robert Johne Jan-Hendrik Klein Steffen Ihlenfeldt Quality Prediction and Classification of Process Parameterization for Multi-Material Jetting by Means of Computer Vision and Machine Learning Journal of Manufacturing and Materials Processing additive manufacturing multi-material jetting process monitoring artificial intelligence machine learning computer vision |
| title | Quality Prediction and Classification of Process Parameterization for Multi-Material Jetting by Means of Computer Vision and Machine Learning |
| title_full | Quality Prediction and Classification of Process Parameterization for Multi-Material Jetting by Means of Computer Vision and Machine Learning |
| title_fullStr | Quality Prediction and Classification of Process Parameterization for Multi-Material Jetting by Means of Computer Vision and Machine Learning |
| title_full_unstemmed | Quality Prediction and Classification of Process Parameterization for Multi-Material Jetting by Means of Computer Vision and Machine Learning |
| title_short | Quality Prediction and Classification of Process Parameterization for Multi-Material Jetting by Means of Computer Vision and Machine Learning |
| title_sort | quality prediction and classification of process parameterization for multi material jetting by means of computer vision and machine learning |
| topic | additive manufacturing multi-material jetting process monitoring artificial intelligence machine learning computer vision |
| url | https://www.mdpi.com/2504-4494/8/1/8 |
| work_keys_str_mv | AT arminreckert qualitypredictionandclassificationofprocessparameterizationformultimaterialjettingbymeansofcomputervisionandmachinelearning AT valentinlang qualitypredictionandclassificationofprocessparameterizationformultimaterialjettingbymeansofcomputervisionandmachinelearning AT stevenweingarten qualitypredictionandclassificationofprocessparameterizationformultimaterialjettingbymeansofcomputervisionandmachinelearning AT robertjohne qualitypredictionandclassificationofprocessparameterizationformultimaterialjettingbymeansofcomputervisionandmachinelearning AT janhendrikklein qualitypredictionandclassificationofprocessparameterizationformultimaterialjettingbymeansofcomputervisionandmachinelearning AT steffenihlenfeldt qualitypredictionandclassificationofprocessparameterizationformultimaterialjettingbymeansofcomputervisionandmachinelearning |