Dynamics of complex-valued fractional-order neural networks

被引：58

作者：

Kaslik, Eva ^{[1
,2
]}

Radulescu, Ileana Rodica ^{[1
,3
]}

机构：

[1] Inst E Austria Timisoara, Bd V Parvan 4,Room 045B, Timisoara 300223, Romania

[2] West Univ Timisoara, Dept Math & Comp Sci, Timisoara, Romania

[3] Univ Politehn Bucuresti, Fac Sci Appl, Bucharest, Romania

来源：

NEURAL NETWORKS | 2017年 / 89卷

关键词：

Neural networks; Fractional order; Stability; Hopf bifurcation; Ring; Hub; STABILITY ANALYSIS; TIME DELAYS; DIFFERENTIAL-EQUATIONS; NONLINEAR-WAVES; NEURONS; RING; MODEL; BIFURCATION; DIFFUSION; SYSTEMS;

D O I：

10.1016/j.neunet.2017.02.011

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The dynamics of complex-valued fractional-order neuronal networks are investigated, focusing on stability, instability and Hopf bifurcations. Sufficient conditions for the asymptotic stability and instability of a steady state of the network are derived, based on the complex system parameters and the fractional order of the system, considering simplified neuronal connectivity structures (hub and ring). In some specific cases, it is possible to identify the critical values of the fractional order for which Hopf bifurcations may occur. Numerical simulations are presented to illustrate the theoretical findings and to investigate the stability of the limit cycles which appear due to Hopf bifurcations. (C) 2017 Elsevier Ltd. All rights reserved.

引用

页码：39 / 49

页数：11

共 77 条

[1] Multilayer Neural Network with Multi-Valued Neurons in time series forecasting of oil production [J].