2011 Fiscal Year Final Research Report
Model Reduction for High-dimensional Nonlinear Chaotic Dynamical Systems Based on Kernel Multivariate Analysis
Project/Area Number |
22740258
|
Research Category |
Grant-in-Aid for Young Scientists (B)
|
Allocation Type | Single-year Grants |
Research Field |
Mathematical physics/Fundamental condensed matter physics
|
Research Institution | Kagoshima University |
Principal Investigator |
|
Project Period (FY) |
2010 – 2011
|
Keywords | 非平衡 / 非線形物理学 / 時系列解析 / カーネル法 / 多様体学習 / カオス / データ同化 / 水滴落下 |
Research Abstract |
We improved ISOMAP, which is a representative manifold learning method, by preventing mistakes in estimating geodesics between points in a manifold using RANSAC. We applied the proposed approach to the numerical simulation data generated from a mass-spring model for dripping faucet in order to demonstrate the usefulness. Furthermore, we tried to analyze real data of the dripping faucet experiment and obtained promising results.
|
Research Products
(5 results)