2010 Fiscal Year Final Research Report
Prediction and Analysis of Time-Series Data by Using Nonlinear Oscillation Circuit Network
| Project/Area Number |
19760270
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Communication/Network engineering
|
|---|
| Research Institution | Hiroshima Institute of Technology |
|---|
Principal Investigator |
YAMAUCHI Masayuki Hiroshima Institute of Technology, 工学部, 准教授 (40384169)
|
|---|
| Project Period (FY) |
2007 – 2010
|
| Keywords | ネットワーク型発振器 / データマイニング / 同期現象 / 時系列データ / 予測 |
|---|
| Research Abstract |
In this study, we suggest a prediction system by using networks of coupled oscillators on which actual circuit are based. These circuits are 3x3 van der Pol oscillators, which are coupled by inductors as a lattice, and some van der Pol oscillators and one resonator, which are coupled mutual inductances between inductors of oscillators and inductors of a resonator. We predict chaos time-series data and air temperature time-series data by Pspice many using elements of actual scale and by normalize equations of 3x3 van der Pol oscillator array and Runge-Kutta method.
|
