1999 Fiscal Year Final Research Report Summary
Adaptive Stabilization by Chaos in an Emergence System
| Project/Area Number |
09650493
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
計測・制御工学
|
|---|
| Research Institution | Konan University |
|---|
Principal Investigator |
HIRAI Kazumasa Konan University, Faculty of Science, Professor, 理学部, 教授 (30031058)
|
|---|
| Project Period (FY) |
1997 – 1999
|
| Keywords | Chaos / Emergence system / Compound system / Multi-agent system / Adaptive stabilization |
|---|
| Research Abstract |
In this research project nonlinear phenomena in a complex system whose subsystems are chaotic and are connected by constant gains are investigated and, based in these results, the investigation of adaptive stabilization by chaos in an emergence system are made. The investigation can be achieved by solving the following problems.
1. The occurrence mechanism of chaos : The parameter regions of the occurrence of chaos can be shown for nonlinear system of arbitrary order with cubic characteristic. This method can be applied to a system with time lag and the parameter regions of the occurrence of chaos can be made clear.
2. The occurrence mechanism of emergence function : A nonlinear compound system whose subsystems are continuous or discrete chaotic system is considered. By changing the magnitude of coupled gain, the strength of nonlinearity, and the initial values, it was shown that stabilization, synchronization, wandering-chaos phenomenon and other interesting nonlinear phenomena occur in this system. The occurrence mechanism of these phenomena is also analyzed, which gives a hint to the occurrence mechanism of emergence function.
3. Adaptive stabilization : In a ecological or biological system it can be often seen that an equilibrium state of a subsystem is changed to another state by adaptation. Such adaptive stabilization can be realized in a complex system whose subsystem is represented by a chaotic first order discrete-time system. It can be made clear that the each state of subsystem can be controlled to move to an another state autonomously and it stabilizes in the another state. This control can be made by the subsystem itself by knowing the state of other subsystems one step before.
4. The future aspect-construction of a multi-agent system : The results obtained by this research project can be applied to construction of a multi-agent system, which can be seen in social, economical, and computer network system.
|
Research Products
(12 results)
