서브메뉴
검색
Algorithmic Detection and Statistical Analyses of Plasmoids in Multiple-X-Line Collisionless Magnetotail Reconnection
Algorithmic Detection and Statistical Analyses of Plasmoids in Multiple-X-Line Collisionless Magnetotail Reconnection
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202103001
- ISBN
- 9798280749504
- DDC
- 530
- 서명/저자
- Algorithmic Detection and Statistical Analyses of Plasmoids in Multiple-X-Line Collisionless Magnetotail Reconnection
- 발행사항
- [Sl] : Princeton University, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 197 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-12, Section: B.
- 주기사항
- Advisor: Ji, Hantao.
- 학위논문주기
- Thesis (Ph.D.)--Princeton University, 2025.
- 초록/해제
- 요약Correctly identifying structures during multiple-X-line reconnection is crucial for understanding the couplings of the microscale to the macroscale, such as the potential role that the plasmoid instability plays in reconnection dynamics. This is particularly relevant to studying reconnection regions via analyzing in-situ spacecraft data. One spacecraft traces a 1D path through the 3D plasma. This limitation makes detection of dynamic plasma structures difficult. As such, this work focuses on the developing of algorithms which detect plasmoids in in-situ magnetotail data, and the physics resulting from them.We first develop and use a hand-tuned algorithm. We use measurements from the Magnetospheric Multiscale (MMS) mission to perform the first statistical study of magnetic structures and associated energy dissipation observed during a single period of turbulent magnetic reconnection. The size of the plasmoids and other structures forms a decaying exponential distribution. The magnetic structures are locations of significant energy dissipation via parallel electric field, while dissipation via perpendicular electric field dominates outside of the structures. Significant energy also returns from particles to fields.We next present a method for creating spacecraft-like data which can be used to train Machine Learning (ML) models to detect plasmoids in in-situ magnetotail data. We develop a method for generating "messy" 2D simulation data from which simulated spacecraft trajectories can be constructed. This simulated data is used as training data for ML models intended for use on spacecraft data. The classifier we train is able to detect more than 70% of the plasmoids in the dataset but also has a high false positive rate.We next utilize domain adaptation techniques to adapt our classifier to MMS data. A data pipeline is constructed to process data from the magnetotail. Multiple domain adaptation methods are implemented to construct MMS feature representations that resemble the PIC feature representations that the classifier uses. These are then fed into the classifier to generate predictions. The resulting MMS classifiers are applied to an existing catalog of magnetotail plasmoids. Insights from the models' performance are presented and discussed.
- 일반주제명
- Plasma physics
- 일반주제명
- Statistical physics
- 일반주제명
- Electromagnetics
- 일반주제명
- Aerospace engineering
- 일반주제명
- Computational physics
- 키워드
- Machine Learning
- 키워드
- Spacecraft
- 기타저자
- Princeton University Astrophysical Sciences-Plasma Physics Program
- 기본자료저록
- Dissertations Abstracts International. 86-12B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017356602
■00520260202103001
■006m o d
■007cr#unu||||||||
■020 ▼a9798280749504
■035 ▼a(MiAaPQ)AAI31839767
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a530
■1001 ▼aBergstedt, Kendra A.▼0(orcid)0000-0002-4992-6387
■24510▼aAlgorithmic Detection and Statistical Analyses of Plasmoids in Multiple-X-Line Collisionless Magnetotail Reconnection
■260 ▼a[Sl]▼bPrinceton University▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a197 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-12, Section: B.
■500 ▼aAdvisor: Ji, Hantao.
■5021 ▼aThesis (Ph.D.)--Princeton University, 2025.
■520 ▼aCorrectly identifying structures during multiple-X-line reconnection is crucial for understanding the couplings of the microscale to the macroscale, such as the potential role that the plasmoid instability plays in reconnection dynamics. This is particularly relevant to studying reconnection regions via analyzing in-situ spacecraft data. One spacecraft traces a 1D path through the 3D plasma. This limitation makes detection of dynamic plasma structures difficult. As such, this work focuses on the developing of algorithms which detect plasmoids in in-situ magnetotail data, and the physics resulting from them.We first develop and use a hand-tuned algorithm. We use measurements from the Magnetospheric Multiscale (MMS) mission to perform the first statistical study of magnetic structures and associated energy dissipation observed during a single period of turbulent magnetic reconnection. The size of the plasmoids and other structures forms a decaying exponential distribution. The magnetic structures are locations of significant energy dissipation via parallel electric field, while dissipation via perpendicular electric field dominates outside of the structures. Significant energy also returns from particles to fields.We next present a method for creating spacecraft-like data which can be used to train Machine Learning (ML) models to detect plasmoids in in-situ magnetotail data. We develop a method for generating "messy" 2D simulation data from which simulated spacecraft trajectories can be constructed. This simulated data is used as training data for ML models intended for use on spacecraft data. The classifier we train is able to detect more than 70% of the plasmoids in the dataset but also has a high false positive rate.We next utilize domain adaptation techniques to adapt our classifier to MMS data. A data pipeline is constructed to process data from the magnetotail. Multiple domain adaptation methods are implemented to construct MMS feature representations that resemble the PIC feature representations that the classifier uses. These are then fed into the classifier to generate predictions. The resulting MMS classifiers are applied to an existing catalog of magnetotail plasmoids. Insights from the models' performance are presented and discussed.
■590 ▼aSchool code: 0181.
■650 4▼aPlasma physics
■650 4▼aStatistical physics
■650 4▼aElectromagnetics
■650 4▼aAerospace engineering
■650 4▼aComputational physics
■653 ▼aMagnetotail plasmoids
■653 ▼aMachine Learning
■653 ▼aMagnetospheric Multiscale
■653 ▼aReconnection dynamics
■653 ▼aSpacecraft
■690 ▼a0759
■690 ▼a0800
■690 ▼a0217
■690 ▼a0216
■690 ▼a0607
■690 ▼a0538
■71020▼aPrinceton University▼bAstrophysical Sciences-Plasma Physics Program.
■7730 ▼tDissertations Abstracts International▼g86-12B.
■790 ▼a0181
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17356602▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
