| 作成者 |
|
| 本文言語 |
|
| 出版者 |
|
|
|
| 発行日 |
|
| 収録物名 |
|
| 巻 |
|
| 号 |
|
| 開始ページ |
|
| 終了ページ |
|
| 出版タイプ |
|
| アクセス権 |
|
| JaLC DOI |
|
| 関連DOI |
|
| 関連URI |
|
| 関連情報 |
|
| 概要 |
This paper deals with the application of the Genetic Programming (GP) to modeling of diffusion process by using the DT-CNN (Discrete Time Cellular Neural Network) and its application to the control of... chaos in the cells We try to approximate the dynamics of diffusion process by using the observation of time series based on the GP_ In the GP, the system equations are represented by parse trees and the performance (fitness) of each individual is defined as the inversion of the root mean squane error between the observed data and the output of the system equation. By selecting a pair of individuals having higher fitness, the crossover operation is applied to generate new individuals. Simulation studies for approximating known dynamics by using the observed tune series show good estimation of the systems equations. The condition for the propagation failure of the autowave is also discussed based on the estimated equations It is shown that the estimated threshold value for the propagation failure is close to the result of simulation study. Then, we apply the control method to stabilize the chaotic dynamics in the DT-CNN. In our control, since the system equations are estimated, we only need to change the input so that the system moves to the stable region. By assuming the targeted dynamic system f(x(t)) with input s(t) = 0 is estimated by using the GP (denoted f(x(t))), then we impose the input u(t) so that xf = x(t+1) = ∫(x(t))+s(t) where xj is the fixed point. Then, the next state x(t+1) of targeted dynamic system f(x(t)) is replaced by x(t + 1) + s(t). The approximation and control method are applied to chaotic dynamics in DT-CNNs, and they are lead to a fixed point or limit cycle.続きを見る
|
Mendeley出力
KJ00000704199-00001