Modelling fatigue life prediction of additively manufactured Ti-6Al-4V samples using machine learning approach

Jan Horňas; Jiří Běhal; Petr Homola; Sascha Senck; Martin Holzleitner; Norica Godja; Zsolt Pásztor; Bálint Hegedüs; Radek Doubrava; Roman Růžek; Lucie Petrusová

doi:10.1016/j.ijfatigue.2022.107483

Modelling fatigue life prediction of additively manufactured Ti-6Al-4V samples using machine learning approach

Jan Horňas, Jiří Běhal, Petr Homola, Sascha Senck, Martin Holzleitner, Norica Godja, Zsolt Pásztor, Bálint Hegedüs, Radek Doubrava, Roman Růžek, Lucie Petrusová

Research output: Contribution to journal › Article › peer-review

33 Citations (Scopus)

Abstract

In this work, a framework based on the machine learning (ML) approach and Spearman's rank correlation analysis is introduced as an effective instrument to solve the influence of defects detected by micro-computed tomography (μCT) method, and stress amplitude on the fatigue life performance of AM Ti-6Al-4V. Artificial neural network (ANN), random forest regressor (RFR) and support vector regressor (SVR) models are implemented and optimized. The optimization is performed on training set by tuning the hyperparameters and parameters using the leave-one-out cross validation (LOOCV) technique. The results present comparison between predicted and experimental results and validate the proposed framework.

Original language	English
Article number	107483
Journal	International Journal of Fatigue
Volume	169
DOIs	https://doi.org/10.1016/j.ijfatigue.2022.107483
Publication status	Published - Apr 2023

Keywords

Fatigue life prediction
Machine learning (ML)
Micro-computed tomography (µCT)
Selective laser melting (SLM)
Ti-6Al-4V

Access to Document

10.1016/j.ijfatigue.2022.107483

Cite this

Horňas, J., Běhal, J., Homola, P., Senck, S., Holzleitner, M., Godja, N., Pásztor, Z., Hegedüs, B., Doubrava, R., Růžek, R., & Petrusová, L. (2023). Modelling fatigue life prediction of additively manufactured Ti-6Al-4V samples using machine learning approach. International Journal of Fatigue, 169, Article 107483. https://doi.org/10.1016/j.ijfatigue.2022.107483

@article{2b40b36155d74bd09b85d7f67f4aac14,

title = "Modelling fatigue life prediction of additively manufactured Ti-6Al-4V samples using machine learning approach",

abstract = "In this work, a framework based on the machine learning (ML) approach and Spearman's rank correlation analysis is introduced as an effective instrument to solve the influence of defects detected by micro-computed tomography (μCT) method, and stress amplitude on the fatigue life performance of AM Ti-6Al-4V. Artificial neural network (ANN), random forest regressor (RFR) and support vector regressor (SVR) models are implemented and optimized. The optimization is performed on training set by tuning the hyperparameters and parameters using the leave-one-out cross validation (LOOCV) technique. The results present comparison between predicted and experimental results and validate the proposed framework.",

keywords = "Fatigue life prediction, Machine learning (ML), Micro-computed tomography (µCT), Selective laser melting (SLM), Ti-6Al-4V",

author = "Jan Hor{\v n}as and Ji{\v r}{\'i} B{\v e}hal and Petr Homola and Sascha Senck and Martin Holzleitner and Norica Godja and Zsolt P{\'a}sztor and B{\'a}lint Heged{\"u}s and Radek Doubrava and Roman Rů{\v z}ek and Lucie Petrusov{\'a}",

note = "Funding Information: The research leading to this work has received funding from the European Community{\textquoteright}s H2020 for the Clean Sky joint technology initiative under grant agreement No. 101007830 and from the Ministry of Industry and Trade of the Czech Republic in the DKRV01 program dedicated to the development of research organizations. Funding Information: The research leading to this work has received funding from the European Community's H2020 for the Clean Sky joint technology initiative under grant agreement No. 101007830 and from the Ministry of Industry and Trade of the Czech Republic in the DKRV01 program dedicated to the development of research organizations. Publisher Copyright: {\textcopyright} 2022",

year = "2023",

month = apr,

doi = "10.1016/j.ijfatigue.2022.107483",

language = "English",

volume = "169",

journal = "International Journal of Fatigue",

issn = "0142-1123",

publisher = "Elsevier",

}

TY - JOUR

T1 - Modelling fatigue life prediction of additively manufactured Ti-6Al-4V samples using machine learning approach

AU - Horňas, Jan

AU - Běhal, Jiří

AU - Homola, Petr

AU - Senck, Sascha

AU - Holzleitner, Martin

AU - Godja, Norica

AU - Pásztor, Zsolt

AU - Hegedüs, Bálint

AU - Doubrava, Radek

AU - Růžek, Roman

AU - Petrusová, Lucie

N1 - Funding Information: The research leading to this work has received funding from the European Community’s H2020 for the Clean Sky joint technology initiative under grant agreement No. 101007830 and from the Ministry of Industry and Trade of the Czech Republic in the DKRV01 program dedicated to the development of research organizations. Funding Information: The research leading to this work has received funding from the European Community's H2020 for the Clean Sky joint technology initiative under grant agreement No. 101007830 and from the Ministry of Industry and Trade of the Czech Republic in the DKRV01 program dedicated to the development of research organizations. Publisher Copyright: © 2022

PY - 2023/4

Y1 - 2023/4

N2 - In this work, a framework based on the machine learning (ML) approach and Spearman's rank correlation analysis is introduced as an effective instrument to solve the influence of defects detected by micro-computed tomography (μCT) method, and stress amplitude on the fatigue life performance of AM Ti-6Al-4V. Artificial neural network (ANN), random forest regressor (RFR) and support vector regressor (SVR) models are implemented and optimized. The optimization is performed on training set by tuning the hyperparameters and parameters using the leave-one-out cross validation (LOOCV) technique. The results present comparison between predicted and experimental results and validate the proposed framework.

AB - In this work, a framework based on the machine learning (ML) approach and Spearman's rank correlation analysis is introduced as an effective instrument to solve the influence of defects detected by micro-computed tomography (μCT) method, and stress amplitude on the fatigue life performance of AM Ti-6Al-4V. Artificial neural network (ANN), random forest regressor (RFR) and support vector regressor (SVR) models are implemented and optimized. The optimization is performed on training set by tuning the hyperparameters and parameters using the leave-one-out cross validation (LOOCV) technique. The results present comparison between predicted and experimental results and validate the proposed framework.

KW - Fatigue life prediction

KW - Machine learning (ML)

KW - Micro-computed tomography (µCT)

KW - Selective laser melting (SLM)

KW - Ti-6Al-4V

UR - http://www.scopus.com/inward/record.url?scp=85146099044&partnerID=8YFLogxK

U2 - 10.1016/j.ijfatigue.2022.107483

DO - 10.1016/j.ijfatigue.2022.107483

M3 - Article

AN - SCOPUS:85146099044

SN - 0142-1123

VL - 169

JO - International Journal of Fatigue

JF - International Journal of Fatigue

M1 - 107483

ER -

Modelling fatigue life prediction of additively manufactured Ti-6Al-4V samples using machine learning approach

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this