View : 35 Download: 0

Machine Learning Analysis for the Soliton Formation in Resonant Nonlinear Three-Wave Interactions

Title
Machine Learning Analysis for the Soliton Formation in Resonant Nonlinear Three-Wave Interactions
Authors
Kim, Yeun JungLee, MinsooLee, Hae June
Ewha Authors
이민수
SCOPUS Author ID
이민수scopus
Issue Date
2019
Journal Title
JOURNAL OF THE KOREAN PHYSICAL SOCIETY
ISSN
0374-4884JCR Link

1976-8524JCR Link
Citation
JOURNAL OF THE KOREAN PHYSICAL SOCIETY vol. 75, no. 11, pp. 909 - 916
Keywords
SolitonsRaman backscatteringMachine learning
Publisher
KOREAN PHYSICAL SOC
Indexed
SCI; SCIE; SCOPUS; KCI WOS scopus
Document Type
Article
Abstract
For the prediction of nonlinear phenomena in a three-wave Raman backscattering for laser amplification, a machine learning technology is applied to predict the generation of solitons in complicated multi-dimensional parameter spaces. The generation of the soliton in the resonant three-wave system is simulated with one-dimensional fluid equations. The solitons are generated in the early phase of the three-wave interaction, and the slow propagation speeds play an important role. Using a pattern matching method comparing the simulation data with the analytic solution, the generation of solitons are automatically detected. After collecting enough data sets by autonomous parameter scanning in the numerical simulation, nonlinear regression and k-nearest neighbor algorithms are utilized for the prediction of the existence of solitons.
DOI
10.3938/jkps.75.909
Appears in Collections:
엘텍공과대학 > 컴퓨터공학과 > Journal papers
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE