國立虎尾科技大學 |

Differentiable Programming for Computational Plasma Physics.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Differentiable Programming for Computational Plasma Physics./
Author:	McGreivy, Nicholas Bradley.
Description:	1 online resource (355 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
Contained By:	Dissertations Abstracts International85-12B.
Subject:	Computer science. -
Online resource:	click for full text (PQDT)
ISBN:	9798382810317

Differentiable Programming for Computational Plasma Physics.
McGreivy, Nicholas Bradley.

Differentiable Programming for Computational Plasma Physics. - 1 online resource (355 pages)

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

Thesis (Ph.D.)--Princeton University, 2024.

Includes bibliographical references

Differentiable programming allows for derivatives of functions implemented via computer code to be calculated automatically. These derivatives are calculated using automatic differentiation (AD). This thesis explores two applications of differentiable programming to computational plasma physics. First, we consider how differentiable programming can be used to simplify and improve stellarator optimization. We introduce a stellarator coil design code (FOCUSADD) that uses gradient-based optimization to produce stellarator coils with finite build. Because we use reverse mode AD, which can compute gradients of scalar functions with the same computational complexity as the function, FOCUSADD is simple, flexible, and efficient. We then discuss two additional applications of AD in stellarator optimization: finding non-axisymmetric magnetic fields that satisfy magnetohydrodynamic (MHD) equilibrium, and optimizing those magnetic fields subject to MHD equilibrium. Second, we explore how machine learning (ML) can be used to improve or replace the numerical methods used to solve partial differential equations (PDEs), focusing on time-dependent PDEs in fluid mechanics relevant to plasma physics. Differentiable programming allows neural networks and other techniques from ML to be embedded within numerical methods. This is a promising, but relatively new, research area. We focus on two basic questions. First, can we design ML-based PDE solvers that have the same guarantees of conservation, stability, and positivity that standard numerical methods do? The answer is yes; we introduce error-correcting algorithms that preserve invariants of time-dependent PDEs. Second, which types of ML-based solvers work best at solving PDEs? We perform a systematic review of the scientific literature on solving PDEs with ML. Unfortunately, we discover two issues, weak baselines and reporting biases, that affect the interpretation reproducibility of a significant majority of published research. We conclude that using ML to solve PDEs is not as promising as we initially believed.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798382810317Subjects--Topical Terms:

573171
Computer science.
Subjects--Index Terms:

Automatic differentiationIndex Terms--Genre/Form:

554714
Electronic books.

Differentiable Programming for Computational Plasma Physics.
LDR:03454ntm a22003977 4500 001 1151280
005 20241104055845.5
006 m o d
007 cr bn ---uuuuu
008 250605s2024 xx obm 000 0 eng d
020 $a 9798382810317
035 $a (MiAaPQ)AAI31294774
035 $a AAI31294774
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a McGreivy, Nicholas Bradley. $3 1478014
245 1 0 $a Differentiable Programming for Computational Plasma Physics.
264 0 $c 2024
300 $a 1 online resource (355 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 Advisor: Hakim, Ammar H.
502 $a Thesis (Ph.D.)--Princeton University, 2024.
504 $a Includes bibliographical references
520 $a Differentiable programming allows for derivatives of functions implemented via computer code to be calculated automatically. These derivatives are calculated using automatic differentiation (AD). This thesis explores two applications of differentiable programming to computational plasma physics. First, we consider how differentiable programming can be used to simplify and improve stellarator optimization. We introduce a stellarator coil design code (FOCUSADD) that uses gradient-based optimization to produce stellarator coils with finite build. Because we use reverse mode AD, which can compute gradients of scalar functions with the same computational complexity as the function, FOCUSADD is simple, flexible, and efficient. We then discuss two additional applications of AD in stellarator optimization: finding non-axisymmetric magnetic fields that satisfy magnetohydrodynamic (MHD) equilibrium, and optimizing those magnetic fields subject to MHD equilibrium. Second, we explore how machine learning (ML) can be used to improve or replace the numerical methods used to solve partial differential equations (PDEs), focusing on time-dependent PDEs in fluid mechanics relevant to plasma physics. Differentiable programming allows neural networks and other techniques from ML to be embedded within numerical methods. This is a promising, but relatively new, research area. We focus on two basic questions. First, can we design ML-based PDE solvers that have the same guarantees of conservation, stability, and positivity that standard numerical methods do? The answer is yes; we introduce error-correcting algorithms that preserve invariants of time-dependent PDEs. Second, which types of ML-based solvers work best at solving PDEs? We perform a systematic review of the scientific literature on solving PDEs with ML. Unfortunately, we discover two issues, weak baselines and reporting biases, that affect the interpretation reproducibility of a significant majority of published research. We conclude that using ML to solve PDEs is not as promising as we initially believed.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Computational physics. $3 1181955
650 4 $a Plasma physics. $2 bicssc $3 1030958
653 $a Automatic differentiation
653 $a Machine learning
653 $a Partial differential equations
653 $a Stellarator optimization
655 7 $a Electronic books. $2 local $3 554714
690 $a 0759
690 $a 0216
690 $a 0984
690 $a 0800
710 2 $a Princeton University. $b Astrophysical Sciences-Plasma Physics Program. $3 1477926
710 2 $a ProQuest Information and Learning Co. $3 1178819
773 0 $t Dissertations Abstracts International $g 85-12B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31294774 $z click for full text (PQDT)