Hot Flow Curve Description of CuFe2 Alloy via Different Artificial Neural Network Approaches

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27360%2F19%3A10242766" target="_blank" >RIV/61989100:27360/19:10242766 - isvavai.cz</a>

Výsledek na webu

<a href="https://link.springer.com/article/10.1007/s11665-019-04199-5?wt_mc=Internal.Event.1.SEM.ArticleAuthorAssignedToIssue&utm_source=ArticleAuthorAssignedToIssue&utm_medium=email&utm_content=AA_en_06082018&ArticleAuthorAssignedToIssue_20190902" target="_blank" >https://link.springer.com/article/10.1007/s11665-019-04199-5?wt_mc=Internal.Event.1.SEM.ArticleAuthorAssignedToIssue&utm_source=ArticleAuthorAssignedToIssue&utm_medium=email&utm_content=AA_en_06082018&ArticleAuthorAssignedToIssue_20190902</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/s11665-019-04199-5" target="_blank" >10.1007/s11665-019-04199-5</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Hot Flow Curve Description of CuFe2 Alloy via Different Artificial Neural Network Approaches

Popis výsledku v původním jazyce

In this research, a mathematical description of hot flow curves of CuFe2 copper alloy has been assembled. Experimental flow curves of the investigated alloy were created on the basis of hot compression dataset. This dataset was acquired in the temperature range of 923-1223 K and the strain rate range of 0.1-10 s(-1), with the maximum true strain value of 1.0. The experimental flow curves were described by two artificial neural network approaches. In the first case, a neural network has been created to approximate the experimental flow curves with respect to the true strain, strain rate and temperature. In the second case, a hybrid approach based on the combination of predictive models with neural networks has been utilized. In this case, five neural networks were used to describe parameters of these models in relation to the temperature and strain rate. Results have shown that the hybrid approach allows an accurate description of the experimental data and also provides more reliable prediction.

Název v anglickém jazyce

Hot Flow Curve Description of CuFe2 Alloy via Different Artificial Neural Network Approaches

Popis výsledku anglicky

In this research, a mathematical description of hot flow curves of CuFe2 copper alloy has been assembled. Experimental flow curves of the investigated alloy were created on the basis of hot compression dataset. This dataset was acquired in the temperature range of 923-1223 K and the strain rate range of 0.1-10 s(-1), with the maximum true strain value of 1.0. The experimental flow curves were described by two artificial neural network approaches. In the first case, a neural network has been created to approximate the experimental flow curves with respect to the true strain, strain rate and temperature. In the second case, a hybrid approach based on the combination of predictive models with neural networks has been utilized. In this case, five neural networks were used to describe parameters of these models in relation to the temperature and strain rate. Results have shown that the hybrid approach allows an accurate description of the experimental data and also provides more reliable prediction.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

20500 - Materials engineering

Projekt

<a href="/cs/project/EF17_049%2F0008399" target="_blank" >EF17_049/0008399: Rozvoj mezisektorové spolupráce RMTVC s aplikační sférou v oblasti výzkumu progresivních a inovací klasických kovových materiálů a technologií s využitím metod modelování</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2019

Kód důvěrnosti údajů

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

Název periodika

Journal of Materials Engineering and Performance

ISSN

1059-9495

e-ISSN

—

Svazek periodika

28

Číslo periodika v rámci svazku

8

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

8

Strana od-do

4863-4870

Kód UT WoS článku

000483700500033

EID výsledku v databázi Scopus

—