Mechanistic machine learning for metamaterial fatigue strength design from first principles in additive manufacturing
Digital control in manufacturing processes produces significant amounts of metadata. The production process metadata, such as thermal and optical measurements, enables a higher degree of property grading than uninstrumented manufacturing and feedback for fault detection. This study explores how meta...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-05-01
|
| Series: | Materials & Design |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0264127524002624 |
| _version_ | 1797213494023553024 |
|---|---|
| author | Mustafa Awd Lobna Saeed Sebastian Münstermann Matthias Faes Frank Walther |
| author_facet | Mustafa Awd Lobna Saeed Sebastian Münstermann Matthias Faes Frank Walther |
| author_sort | Mustafa Awd |
| collection | DOAJ |
| description | Digital control in manufacturing processes produces significant amounts of metadata. The production process metadata, such as thermal and optical measurements, enables a higher degree of property grading than uninstrumented manufacturing and feedback for fault detection. This study explores how metadata can design fatigue-resistant structures using physically grounded models such as density functional theory, cyclic plasticity, and fracture mechanics that train machine learning algorithms. Machine learning models work very efficiently in their trained physical space. In comparison, mechanistic models are computationally costly for complex phenomena such as fatigue. We show how fatigue can be administered consistently at all scales by energy-based criteria and how a mechanistic function can be built based on this concept. The energy mechanistic function allows exact quantification of the effect of the existing flaws from manufacturing on fatigue lifetime under certain load boundary conditions. Since the mechanistic function is local and subscale to the prediction scale of the machine learning model, it can be used to build density functions for probabilistic regression of the fatigue property on the scale above. The analysis is applied to the selective laser melting process due to the availability of digital control and metadata generation during deposition. |
| first_indexed | 2024-04-24T10:59:10Z |
| format | Article |
| id | doaj.art-ee0b2b66fcf048f7899f10d759c52cea |
| institution | Directory Open Access Journal |
| issn | 0264-1275 |
| language | English |
| last_indexed | 2024-04-24T10:59:10Z |
| publishDate | 2024-05-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Materials & Design |
| spelling | doaj.art-ee0b2b66fcf048f7899f10d759c52cea2024-04-12T04:44:17ZengElsevierMaterials & Design0264-12752024-05-01241112889Mechanistic machine learning for metamaterial fatigue strength design from first principles in additive manufacturingMustafa Awd0Lobna Saeed1Sebastian Münstermann2Matthias Faes3Frank Walther4Chair of Materials Test Engineering (WPT), TU Dortmund University, Baroper Str. 303, Dortmund, D-44227, Germany; Corresponding author.Crystallography and Geomaterials Research Group, Faculty of Geosciences, University of Bremen, Klagenfurter Str. 2-4, Bremen, D-28359, GermanyInstitute for Metal Forming, RWTH Aachen University, Intzestr. 10, Aachen, D-52072, GermanyChair for Reliability Engineering (CRE), TU Dortmund University, Leonhard-Euler Str. 5, Dortmund, D-44227, GermanyChair of Materials Test Engineering (WPT), TU Dortmund University, Baroper Str. 303, Dortmund, D-44227, GermanyDigital control in manufacturing processes produces significant amounts of metadata. The production process metadata, such as thermal and optical measurements, enables a higher degree of property grading than uninstrumented manufacturing and feedback for fault detection. This study explores how metadata can design fatigue-resistant structures using physically grounded models such as density functional theory, cyclic plasticity, and fracture mechanics that train machine learning algorithms. Machine learning models work very efficiently in their trained physical space. In comparison, mechanistic models are computationally costly for complex phenomena such as fatigue. We show how fatigue can be administered consistently at all scales by energy-based criteria and how a mechanistic function can be built based on this concept. The energy mechanistic function allows exact quantification of the effect of the existing flaws from manufacturing on fatigue lifetime under certain load boundary conditions. Since the mechanistic function is local and subscale to the prediction scale of the machine learning model, it can be used to build density functions for probabilistic regression of the fatigue property on the scale above. The analysis is applied to the selective laser melting process due to the availability of digital control and metadata generation during deposition.http://www.sciencedirect.com/science/article/pii/S0264127524002624Density functional theory (DFT)Functional gradingMachine learning (ML)Fatigue strengthCohesion energyAdditive manufacturing (AM) |
| spellingShingle | Mustafa Awd Lobna Saeed Sebastian Münstermann Matthias Faes Frank Walther Mechanistic machine learning for metamaterial fatigue strength design from first principles in additive manufacturing Materials & Design Density functional theory (DFT) Functional grading Machine learning (ML) Fatigue strength Cohesion energy Additive manufacturing (AM) |
| title | Mechanistic machine learning for metamaterial fatigue strength design from first principles in additive manufacturing |
| title_full | Mechanistic machine learning for metamaterial fatigue strength design from first principles in additive manufacturing |
| title_fullStr | Mechanistic machine learning for metamaterial fatigue strength design from first principles in additive manufacturing |
| title_full_unstemmed | Mechanistic machine learning for metamaterial fatigue strength design from first principles in additive manufacturing |
| title_short | Mechanistic machine learning for metamaterial fatigue strength design from first principles in additive manufacturing |
| title_sort | mechanistic machine learning for metamaterial fatigue strength design from first principles in additive manufacturing |
| topic | Density functional theory (DFT) Functional grading Machine learning (ML) Fatigue strength Cohesion energy Additive manufacturing (AM) |
| url | http://www.sciencedirect.com/science/article/pii/S0264127524002624 |
| work_keys_str_mv | AT mustafaawd mechanisticmachinelearningformetamaterialfatiguestrengthdesignfromfirstprinciplesinadditivemanufacturing AT lobnasaeed mechanisticmachinelearningformetamaterialfatiguestrengthdesignfromfirstprinciplesinadditivemanufacturing AT sebastianmunstermann mechanisticmachinelearningformetamaterialfatiguestrengthdesignfromfirstprinciplesinadditivemanufacturing AT matthiasfaes mechanisticmachinelearningformetamaterialfatiguestrengthdesignfromfirstprinciplesinadditivemanufacturing AT frankwalther mechanisticmachinelearningformetamaterialfatiguestrengthdesignfromfirstprinciplesinadditivemanufacturing |