東華大學圖書館 |

A Hybrid Data-Driven Methodology to Determine Additive Manufacturing Process Parameters /

Record Type:	Electronic resources : Monograph/item
Title/Author:	A Hybrid Data-Driven Methodology to Determine Additive Manufacturing Process Parameters // Angshuman Deka.
Author:	Deka, Angshuman,
Description:	1 electronic resource (198 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
Contained By:	Dissertations Abstracts International85-12B.
Subject:	Mechanical engineering. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31294397
ISBN:	9798382831336

A Hybrid Data-Driven Methodology to Determine Additive Manufacturing Process Parameters /
Deka, Angshuman,

A Hybrid Data-Driven Methodology to Determine Additive Manufacturing Process Parameters /Angshuman Deka. - 1 electronic resource (198 pages)

Source: Dissertations Abstracts International, Volume: 85-12, Section: B.

Additive manufacturing (AM) has undergone significant advancements over the years offering advantages not found in conventional manufacturing processes. AM enables direct part production from digital models, facilitates the possibility of multi-material complex shapes without requiring assembly, and reduces the new product development cycle time. Despite these benefits, challenges persist in adopting AM as a full-scale production technology for manufacturing and mass production. One major obstacle in its adoption is the accurate identification of material and processing parameters to achieve the desired end properties of AM products.This study addresses the challenge specific to the material extrusion process, trademarked Fused Deposition Modeling (FDM) by Stratasys, focusing on determining manufacturing parameters to achieve the desired product properties. In FDM, the properties of the end product are influenced by both the filament material and the FDM processing or machine parameters, including extrusion speed, extrusion nozzle temperature, layer thickness, and print speed. Due to the complex physical phenomena of FDM resulting in interactions among the material and processing parameters, efficiently predicting the parameters required to achieve desired properties proves very challenging. Typically, a trial-and-error experimental approach is used to determine manufacturing parameters for required FDM product properties and for establishing the relationships among the manufacturing parameters and properties. This process involves printing and testing multiple samples and is time-consuming and expensive. Hence, there is a need for efficient, knowledge-based engineering design methods to assist in making design decisions or selecting manufacturing parameters to achieve desired product properties that are effective and useful in case of limited information and varying manufacturing conditions.To address these challenges, this work develops data-driven computational design methods for efficient design decision-making and parameter identification for both material and processing parameters in FDM. Considering that both material and process contribute to product properties, computational design methods are established to make design decisions or determine design parameters for filament material and processing parameters in FDM to achieve desired product properties. Therefore, this study implements integrated material and process design using a target oriented inverse design method, incorporating both conventional and additive manufacturing to achieve desired product properties. Additionally, the research addresses the multifaceted challenges associated with implementing such computational design methods. This work thus addresses design challenges when limited process information or data is available, by implementing a novel predictive modeling-based method for simulating necessary data to make efficient decisions for achieving product properties in the FDM process. Next, the adaptability and flexibility of this design method is established for addressing design decisions under varying manufacturing conditions or design requirements. Finally, this research examines the uncertainty associated with the computational design and manufacturing process to comprehend the product property variation and for designing products that are resilient to such variations. The efficacy of the generic computational methods is demonstrated through diverse examples further establishing their versatility and effectiveness and thus pave the way for broader implementation for other FDM or AM design problems. The adaptability and robustness of these methods is demonstrated by successfully addressing various design challenges across different scenarios. These examples not only demonstrate the utility of the computational methods developed but also serves as compelling evidence of their potential for broader implementation to address other FDM or AM design problems.

English

ISBN: 9798382831336Subjects--Topical Terms:

649730
Mechanical engineering.
Subjects--Index Terms:

Additive manufacturing

A Hybrid Data-Driven Methodology to Determine Additive Manufacturing Process Parameters /
LDR:05574nmm a22004693i 4500 001 2438589
005 20260727134358.5
006 m o d
007 cr|nu||||||||
008 261011s2024 miu||||||m |||||||eng d
020 $a 9798382831336
035 $a (MiAaPQD)AAI31294397
035 $a AAI31294397
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Deka, Angshuman, $e author. $3 3829945
245 1 2 $a A Hybrid Data-Driven Methodology to Determine Additive Manufacturing Process Parameters / $c Angshuman Deka.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (198 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: Dissertations Abstracts International, Volume: 85-12, Section: B.
500 $a Advisors: Hall, John Committee members: Dargush, Gary F.; Zhou, Chi.
502 $b Ph.D. $c State University of New York at Buffalo $d 2024.
520 $a Additive manufacturing (AM) has undergone significant advancements over the years offering advantages not found in conventional manufacturing processes. AM enables direct part production from digital models, facilitates the possibility of multi-material complex shapes without requiring assembly, and reduces the new product development cycle time. Despite these benefits, challenges persist in adopting AM as a full-scale production technology for manufacturing and mass production. One major obstacle in its adoption is the accurate identification of material and processing parameters to achieve the desired end properties of AM products.This study addresses the challenge specific to the material extrusion process, trademarked Fused Deposition Modeling (FDM) by Stratasys, focusing on determining manufacturing parameters to achieve the desired product properties. In FDM, the properties of the end product are influenced by both the filament material and the FDM processing or machine parameters, including extrusion speed, extrusion nozzle temperature, layer thickness, and print speed. Due to the complex physical phenomena of FDM resulting in interactions among the material and processing parameters, efficiently predicting the parameters required to achieve desired properties proves very challenging. Typically, a trial-and-error experimental approach is used to determine manufacturing parameters for required FDM product properties and for establishing the relationships among the manufacturing parameters and properties. This process involves printing and testing multiple samples and is time-consuming and expensive. Hence, there is a need for efficient, knowledge-based engineering design methods to assist in making design decisions or selecting manufacturing parameters to achieve desired product properties that are effective and useful in case of limited information and varying manufacturing conditions.To address these challenges, this work develops data-driven computational design methods for efficient design decision-making and parameter identification for both material and processing parameters in FDM. Considering that both material and process contribute to product properties, computational design methods are established to make design decisions or determine design parameters for filament material and processing parameters in FDM to achieve desired product properties. Therefore, this study implements integrated material and process design using a target oriented inverse design method, incorporating both conventional and additive manufacturing to achieve desired product properties. Additionally, the research addresses the multifaceted challenges associated with implementing such computational design methods. This work thus addresses design challenges when limited process information or data is available, by implementing a novel predictive modeling-based method for simulating necessary data to make efficient decisions for achieving product properties in the FDM process. Next, the adaptability and flexibility of this design method is established for addressing design decisions under varying manufacturing conditions or design requirements. Finally, this research examines the uncertainty associated with the computational design and manufacturing process to comprehend the product property variation and for designing products that are resilient to such variations. The efficacy of the generic computational methods is demonstrated through diverse examples further establishing their versatility and effectiveness and thus pave the way for broader implementation for other FDM or AM design problems. The adaptability and robustness of these methods is demonstrated by successfully addressing various design challenges across different scenarios. These examples not only demonstrate the utility of the computational methods developed but also serves as compelling evidence of their potential for broader implementation to address other FDM or AM design problems.
546 $a English
590 $a School code: 0656
650 4 $a Mechanical engineering. $3 649730
650 4 $a Industrial engineering. $3 526216
650 4 $a Materials science. $3 543314
650 4 $a Computer science. $3 523869
650 4 $a Mechanics. $3 525881
653 $a Additive manufacturing
653 $a Fused Deposition Modeling
653 $a Hybrid data-driven design
653 $a Predictive modeling
653 $a Response surface methodology
653 $a Uncertainty quantification
690 $a 0548
690 $a 0546
690 $a 0984
690 $a 0346
690 $a 0794
710 2 $a State University of New York at Buffalo. $b Mechanical and Aerospace Engineering. $e degree granting institution. $3 3818574
720 1 $a Hall, John $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 85-12B.
790 $a 0656
791 $a Ph.D.
792 $a 2024
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31294397