Multivariate interpolation and machine learning models for extreme defects-based fatigue life prediction of Ti6Al4V specimens fabricated by SLM

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21340%2F25%3A00382216" target="_blank" >RIV/68407700:21340/25:00382216 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1016/j.engfracmech.2024.110756" target="_blank" >https://doi.org/10.1016/j.engfracmech.2024.110756</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1016/j.engfracmech.2024.110756" target="_blank" >10.1016/j.engfracmech.2024.110756</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Multivariate interpolation and machine learning models for extreme defects-based fatigue life prediction of Ti6Al4V specimens fabricated by SLM

Popis výsledku v původním jazyce

The paper proposes a novel methodology for extreme defects-based fatigue life prediction of Ti6Al4V specimens fabricated by selective laser melting (SLM) technique. The introduced framework is represented by a traditional ranking method based on the three highest values of maximum stress intensity factor (Kmax) with related defect parameters (size, distance from the free surface, compactness and sphericity), training set augmentation using variational autoencoder (VAE) and optimized data-driven models. The defects were observed using micro-computed tomography (mu -CT) prior to the fatigue tests. As data-driven methods a multivariate interpolation and machine learning (ML) models were employed and tuned using Bayesian optimization algorithm called tree-structured Parzen estimator (TPE). The proposed methodology was validated on the test set and the highest prediction accuracy was achieved by random forest (RF) model with value of coefficient of determination R2test = 0.956. Additionally, the SHapley Additive exPlanations (SHAP) analysis was conducted to gain a deeper insights of applied data-driven models.

Název v anglickém jazyce

Multivariate interpolation and machine learning models for extreme defects-based fatigue life prediction of Ti6Al4V specimens fabricated by SLM

Popis výsledku anglicky

The paper proposes a novel methodology for extreme defects-based fatigue life prediction of Ti6Al4V specimens fabricated by selective laser melting (SLM) technique. The introduced framework is represented by a traditional ranking method based on the three highest values of maximum stress intensity factor (Kmax) with related defect parameters (size, distance from the free surface, compactness and sphericity), training set augmentation using variational autoencoder (VAE) and optimized data-driven models. The defects were observed using micro-computed tomography (mu -CT) prior to the fatigue tests. As data-driven methods a multivariate interpolation and machine learning (ML) models were employed and tuned using Bayesian optimization algorithm called tree-structured Parzen estimator (TPE). The proposed methodology was validated on the test set and the highest prediction accuracy was achieved by random forest (RF) model with value of coefficient of determination R2test = 0.956. Additionally, the SHapley Additive exPlanations (SHAP) analysis was conducted to gain a deeper insights of applied data-driven models.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

20501 - Materials engineering

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Engineering Fracture Mechanics

ISSN

0013-7944

e-ISSN

1873-7315

Svazek periodika

314

Číslo periodika v rámci svazku

110756

Stát vydavatele periodika

GB - Spojené království Velké Británie a Severního Irska

Počet stran výsledku

26

Strana od-do

—

Kód UT WoS článku

001421357100001

EID výsledku v databázi Scopus

2-s2.0-85214258699