東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

FindBook

Google Book

Amazon

博客來

Predicting Fatigue Indicator Parameters in Additively Manufactured IN625 Using Genetic Programming for Symbolic Regression.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Predicting Fatigue Indicator Parameters in Additively Manufactured IN625 Using Genetic Programming for Symbolic Regression./
作者:	Hansen, Cooper Kelly.
面頁冊數:	1 online resource (80 pages)
附註:	Source: Masters Abstracts International, Volume: 84-09.
Contained By:	Masters Abstracts International84-09.
標題:	Mechanical engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30243694click for full text (PQDT)
ISBN:	9798374416923

Predicting Fatigue Indicator Parameters in Additively Manufactured IN625 Using Genetic Programming for Symbolic Regression.
Hansen, Cooper Kelly.

Predicting Fatigue Indicator Parameters in Additively Manufactured IN625 Using Genetic Programming for Symbolic Regression. - 1 online resource (80 pages)

Source: Masters Abstracts International, Volume: 84-09.

Thesis (M.S.)--The University of Utah, 2023.

Includes bibliographical references

Fatigue indicator parameters (FIPs) are typically computed using crystal plasticity finite element modeling (CPFEM) and used to predict microscale crack initiation. While informative, computing FIPs in this manner precludes their use in practical fatigue loading cases due to the computational demand from CPFEM. To address this limitation, an interpretable machine learning approach is developed and used to model FIPs in additive manufactured IN625. Accurately predicted FIPs then allow for the prediction of crack initiation and growth in the microstructurally small and physically small regime. Further, these calculations can then be combined with existing models (NASGRO) for long crack growth to then efficiently predict the total fatigue life of the metal.The material of interest, AM IN625, is digitally represented in DREAM3D as a statistical volume element (SVE) and simulated using CPFEM. FIP training data are generated from the CPFEM simulations and corresponding features are the SVE microstructure attributes. Data are then evolved using genetic programming for symbolic regression (GPSR) to achieve the goal of a human-interpretable machine learning (ML) model. A significant feature engineering effort resulted in two important developments: a process of combining features at multiple scales (i.e., grain, subgrain, etc.) to adequately relate the microstructure of an SVE to its FIPs and the utilization of representation learning by way of GPSR to discover complex relationships between potential features and improve the model. As expected, strain quantities proved to be the most highly correlated feature to FIPs. To acquire strain data without conducting complete CPFEM simulations, infinitesimal strain approximations were made that proved efficient and critical to ML model accuracy.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2023

Mode of access: World Wide Web

ISBN: 9798374416923Subjects--Topical Terms:

649730
Mechanical engineering.
Subjects--Index Terms:

Computational mechanicsIndex Terms--Genre/Form:

542853
Electronic books.

Predicting Fatigue Indicator Parameters in Additively Manufactured IN625 Using Genetic Programming for Symbolic Regression.
LDR:03329nmm a2200445K 4500 001 2363158
005 20231116093810.5
006 m o d
007 cr mn ---uuuuu
008 241011s2023 xx obm 000 0 eng d
020 $a 9798374416923
035 $a (MiAaPQ)AAI30243694
035 $a AAI30243694
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Hansen, Cooper Kelly. $3 3703905
245 1 0 $a Predicting Fatigue Indicator Parameters in Additively Manufactured IN625 Using Genetic Programming for Symbolic Regression.
264 0 $c 2023
300 $a 1 online resource (80 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 84-09.
500 $a Advisor: Hochhalter, Jacob Dean.
502 $a Thesis (M.S.)--The University of Utah, 2023.
504 $a Includes bibliographical references
520 $a Fatigue indicator parameters (FIPs) are typically computed using crystal plasticity finite element modeling (CPFEM) and used to predict microscale crack initiation. While informative, computing FIPs in this manner precludes their use in practical fatigue loading cases due to the computational demand from CPFEM. To address this limitation, an interpretable machine learning approach is developed and used to model FIPs in additive manufactured IN625. Accurately predicted FIPs then allow for the prediction of crack initiation and growth in the microstructurally small and physically small regime. Further, these calculations can then be combined with existing models (NASGRO) for long crack growth to then efficiently predict the total fatigue life of the metal.The material of interest, AM IN625, is digitally represented in DREAM3D as a statistical volume element (SVE) and simulated using CPFEM. FIP training data are generated from the CPFEM simulations and corresponding features are the SVE microstructure attributes. Data are then evolved using genetic programming for symbolic regression (GPSR) to achieve the goal of a human-interpretable machine learning (ML) model. A significant feature engineering effort resulted in two important developments: a process of combining features at multiple scales (i.e., grain, subgrain, etc.) to adequately relate the microstructure of an SVE to its FIPs and the utilization of representation learning by way of GPSR to discover complex relationships between potential features and improve the model. As expected, strain quantities proved to be the most highly correlated feature to FIPs. To acquire strain data without conducting complete CPFEM simulations, infinitesimal strain approximations were made that proved efficient and critical to ML model accuracy.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Mechanical engineering. $3 649730
650 4 $a Computational physics. $3 3343998
650 4 $a Mechanics. $3 525881
650 4 $a Materials science. $3 543314
653 $a Computational mechanics
653 $a Crystal plasticity
653 $a Genetic programming
653 $a Machine learning
653 $a Materials science
653 $a Symbolic regression
653 $a Parameters
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0548
690 $a 0216
690 $a 0346
690 $a 0794
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a The University of Utah. $b Mechanical Engineering. $3 1675126
773 0 $t Masters Abstracts International $g 84-09.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30243694 $z click for full text (PQDT)