Publications

Detailed Information

On the emergent behaviors of stochastic flocking models : 확률적 플로킹 모형에서의 창발 현상에 관하여

DC Field Value Language
dc.contributor.advisor하승열-
dc.contributor.author심우주-
dc.date.accessioned2020-10-13T04:02:16Z-
dc.date.available2020-10-13T04:02:16Z-
dc.date.issued2020-
dc.identifier.other000000163063-
dc.identifier.urihttps://hdl.handle.net/10371/170702-
dc.identifier.urihttp://dcollection.snu.ac.kr/common/orgView/000000163063ko_KR
dc.description학위논문 (박사) -- 서울대학교 대학원 : 자연과학대학 수리과학부, 2020. 8. 하승열.-
dc.description.abstractWe introduce the Inertial Spin (IS) model and study its emergent dynamics under various frameworks. We first provide a derivation of Hamiltonian description of the IS model as a three dimensional flocking model with spin, which is an internal variable generating the rotation of velocities. We review some flocking estimates on the IS model and provide several numerical experiments. Then, we formally derive the Justh-Krishnaprasad (J-K) model as a two-dimensional restriction of the IS model under the small inertia regime. For the J-K model, we also review the flocking estimate of the J-K model and present an improved estimate from the previous one. As a mathematical model to describe a flocking behavior in nature, it is natural to assume a randomness on their dynamics. Thus, to provide a better description, one needs to incorporate such uncertain factors to the model and analyze their behaviors on the dynamics and stability of the flocking state. We here provide two different kind of noises on the J-K model and study the corresponding stochastic differential equations to fulfill this. Namely, we considered the update rules of velocity heading angles by adding Gaussian white noise to the update itself or adding the white noise to their coupling strength, which we call the additive and multiplicative noise, respectively. For the additive noise J-K model, we provide a lower-bound estimate on the probability of sample paths of heading angles to be confined in a certain bound in finite time, and also obtain an upper bound of expected order parameter square. For the multiplicative noise J-K model, we show that the multiplicative noise allows the asymptotic alignment of velocity heading angles if the coupling strength is sufficiently large compared to the diffusion.-
dc.description.abstract본 학위 논문에서는, 플로킹 현상을 기술하는 관성 스핀 모형과 그로부터 유도된 확률 모형들에 대하여 연구한다. 먼저 우리는 관성 스핀 모형의 정당화를 위해 3차원 공간에서의 해밀토니언 역학적 관점에 따라 속력을 보존하는 플로킹 모형을 구현하는 방법을 소개하고 그 집단 행동을 분석한다. 그러나 관성 스핀 모형의 집단현상을 실제와 비교함에 있어 입자들의 다이나믹스에 영향을 줄 수 있는 미지의 불확실성을 고려하는 것이 보다 자연스러우므로, 우리는 자연계의 창발 현상을 보다 정확히 설명하기 위해 관성 스핀 모형에 백색소음을 추가하는 여러 가지 방법을 고려한다. 우리가 고려하는 확률 모형은 크게 두 가지인데, 2차원 관성 스핀 모형의 극소 관성 체제 하에서 자연스럽게 유도되는 J-K 모형에 백색 소음을 직접 추가하는 가법적 방법과, J-K 모형의 상호 작용 계수에 백색 소음을 추가하는 곱셈적 방법으로 나뉘어 진다. 가법적 백색소음이 있는 J-K 모형에서는, 주어진 시간 동안 각각의 표본 경로가 플로킹 상태에서 일정 이상 멀어질 수 있는 확률을 어림하고, 또 플로킹 상태에서 가까워짐을 나타내는 질서도의 점근적 상극한을 계산하여 결정론적 J-K 모형과의 차이점을 나타낸다. 그와 반대로 곱셈적 백색소음이 있는 J-K 모형에서는 결정론적 J-K 모형과 같이 플로킹 상태의 점근적 안정성이 발생할 수 있는 충분조건에 대해 공부한다.-
dc.description.tableofcontents1 Introduction 1
2 The Inertial Spin model 8
2.1 A brief review on the IS model 8
2.1.1 Derivation of the IS model 8
2.1.2 Conservation laws 12
2.2 Asymptotic behavior of the IS model 14
2.2.1 Decoupled IS system 14
2.2.2 Emergent dynamics of a many-body system 15
2.3 Numerical simulations 28
2.3.1 Decoupled IS model 28
2.3.2 A coupled IS model 31

3 Justh-Krishnaprasad model with additive noises 36
3.1 Preliminaries 37
3.1.1 From the IS model to the J-K model 38
3.1.2 A brief review on the J-K model 39
3.2 Emergence of flocking for the deterministic J-K model 42
3.3 The stochastic persistency of the additive noise J-K model 49
3.3.1 Basic sample path estimates 49
3.3.2 Relaxed first collision-time 50
3.3.3 Estimate on the relaxed first collision-time 51
3.3.4 Description of main result 57
3.3.5 Proof of Theorem 3.3.1 60
3.4 Order parameter estimate 70
3.5 Numerical simulations 73

4 J-K model with multiplicative noises 75
4.1 Basic properties 76
4.1.1 Derivation of multiplicative noise J-K model 76
4.2 A two-body system 79
4.2.1 ψ-independent noise 80
4.2.2 ψ-dependent noise 83
4.3 A many-body system 86
4.3.1 Many-body system with independent white noises 86
4.3.2 Many-body system with identical white noise 93
4.4 Numerical simulations 99
4.4.1 A two-body system 99
4.4.2 A many-body system 100

5 Conclusion and future works 102

Appendix A Theoretical backgrounds 104
A.1 Barbalat's lemma 104
A.2 Comparison principles for stochastic differential equations 105
A.3 Well-posedness for stochastic differential equations 106
A.4 Strong Markov Property 107

Appendix B The Kuramoto model 109
B.1 Basic descriptions 109
B.2 Previous results 111

Bibliography 113

Abstract (in Korean) 122

Acknowledgement (in Korean) 123
-
dc.language.isoeng-
dc.publisher서울대학교 대학원-
dc.subjectFlocking-
dc.subjectInertial Spin model-
dc.subjectJusth-Krishnaprasad model-
dc.subjectRandom dynamical system-
dc.subjectStochastic differential equation-
dc.subject플로킹-
dc.subject관성 스핀 모델-
dc.subject임의적 동역학계-
dc.subject확률미분방정식-
dc.subject.ddc510-
dc.titleOn the emergent behaviors of stochastic flocking models-
dc.title.alternative확률적 플로킹 모형에서의 창발 현상에 관하여-
dc.typeThesis-
dc.typeDissertation-
dc.contributor.department자연과학대학 수리과학부-
dc.description.degreeDoctor-
dc.date.awarded2020-08-
dc.identifier.uciI804:11032-000000163063-
dc.identifier.holdings000000000043▲000000000048▲000000163063▲-
Appears in Collections:
Files in This Item:

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share