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

Mechanistic Computational Intelligen...

Saha, Sourav,

FindBook

Google Book

Amazon

博客來

Mechanistic Computational Intelligence (MCI) From Theory to Practice: Opening Pathways for a New Generation of Engineering Software /

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Mechanistic Computational Intelligence (MCI) From Theory to Practice: Opening Pathways for a New Generation of Engineering Software // Sourav Saha.
作者:	Saha, Sourav,
面頁冊數:	1 electronic resource (155 pages)
附註:	Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
Contained By:	Dissertations Abstracts International85-11B.
標題:	Mechanics. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31241716
ISBN:	9798382763118

Mechanistic Computational Intelligence (MCI) From Theory to Practice: Opening Pathways for a New Generation of Engineering Software /
Saha, Sourav,

Mechanistic Computational Intelligence (MCI) From Theory to Practice: Opening Pathways for a New Generation of Engineering Software /Sourav Saha. - 1 electronic resource (155 pages)

Source: Dissertations Abstracts International, Volume: 85-11, Section: B.

The field of computational science and engineering is evolving from software 1.0, with explicit, task-specific programming, to software 2.0, which uses data and optimization. However, a direct transition to software 2.0 is extremely difficult for engineering science because a) data is scarce for many engineering systems, b) modeling complex multiphysics problem requires extensive calibration of parameters, and c) augmenting the experimental data with the computational model is extremely time-consuming. A gap exists in the contemporary computational and/or theoretical methods that can engender the next generation of engineering software 2.0. The dissertation proposes a synergistic computational and theoretical framework called Mechanistic Computational Intelligence (MCI) to fill the void. The MCI theory is built upon the mathematical foundations of mechanistic interpolation method and the scientific and engineering principles of Interpolating Neural Network (INN). The MCI aims to solve three general types of problems encountered in science and engineering. The first type of problem occurs when there is no known closed-form physical equation. The second type arises when the knowledge of governing laws is incomplete. Lastly, the third type of problem involves known governing equations where computational speed needs improvement. The MCI provides a solution framework for all three types of problems in unified software system named engineering software 2.0. The envisioned engineering software 2.0 will use a single library of deep learning interpolants that can simultaneously solve/train for a problem. The dissertation defines and discusses the evolution of the MCI theory, provides the details of the building blocks of the MCI, shows how the general form of neural networks and solvers are derived from kernel integral transform, demonstrates the application of MCI theories in approximating mathematical operators and solving the aforementioned three types of problems, and proposes future extensions of the work.

English

ISBN: 9798382763118Subjects--Topical Terms:

525881
Mechanics.
Subjects--Index Terms:

Mechanistic Computational Intelligence

Mechanistic Computational Intelligence (MCI) From Theory to Practice: Opening Pathways for a New Generation of Engineering Software /
LDR:03563nmm a22004333i 4500 001 2429340
005 20250430100803.5
006 m o d
007 cr|nu||||||||
008 261011s2024 miu||||||m |||||||eng d
020 $a 9798382763118
035 $a (MiAaPQD)AAI31241716
035 $a AAI31241716
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Saha, Sourav, $e author. $3 3817130
245 1 0 $a Mechanistic Computational Intelligence (MCI) From Theory to Practice: Opening Pathways for a New Generation of Engineering Software / $c Sourav Saha.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (155 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: 85-11, Section: B.
500 $a Advisors: Liu, Wing Committee members: Wagner, Gregory; Qian, Dong; Bessa, Miguel.
502 $b Ph.D. $c Northwestern University $d 2024.
520 $a The field of computational science and engineering is evolving from software 1.0, with explicit, task-specific programming, to software 2.0, which uses data and optimization. However, a direct transition to software 2.0 is extremely difficult for engineering science because a) data is scarce for many engineering systems, b) modeling complex multiphysics problem requires extensive calibration of parameters, and c) augmenting the experimental data with the computational model is extremely time-consuming. A gap exists in the contemporary computational and/or theoretical methods that can engender the next generation of engineering software 2.0. The dissertation proposes a synergistic computational and theoretical framework called Mechanistic Computational Intelligence (MCI) to fill the void. The MCI theory is built upon the mathematical foundations of mechanistic interpolation method and the scientific and engineering principles of Interpolating Neural Network (INN). The MCI aims to solve three general types of problems encountered in science and engineering. The first type of problem occurs when there is no known closed-form physical equation. The second type arises when the knowledge of governing laws is incomplete. Lastly, the third type of problem involves known governing equations where computational speed needs improvement. The MCI provides a solution framework for all three types of problems in unified software system named engineering software 2.0. The envisioned engineering software 2.0 will use a single library of deep learning interpolants that can simultaneously solve/train for a problem. The dissertation defines and discusses the evolution of the MCI theory, provides the details of the building blocks of the MCI, shows how the general form of neural networks and solvers are derived from kernel integral transform, demonstrates the application of MCI theories in approximating mathematical operators and solving the aforementioned three types of problems, and proposes future extensions of the work.
546 $a English
590 $a School code: 0163
650 4 $a Mechanics. $3 525881
650 4 $a Mechanical engineering. $3 649730
650 4 $a Computational physics. $3 3343998
653 $a Mechanistic Computational Intelligence
653 $a Computational mechanics
653 $a Deep neural network
653 $a Interpolating network
653 $a Mechanistic data science
690 $a 0346
690 $a 0548
690 $a 0216
710 2 $a Northwestern University. $b Theoretical and Applied Mechanics. $e degree granting institution. $3 3817131
720 1 $a Liu, Wing $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 85-11B.
790 $a 0163
791 $a Ph.D.
792 $a 2024
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31241716