Finite-Time Mittag-Leffler Stability of Fractional-Order Quaternion-Valued Memristive Neural Networks with Impulses.

Autor: Pratap, A., Raja, R., Alzabut, J., Dianavinnarasi, J., Cao, J., Rajchakit, G.
Předmět:
Zdroj: Neural Processing Letters; Apr2020, Vol. 51 Issue 2, p1485-1526, 42p
Abstrakt: The finite-time Mittag-Leffler stability for fractional-order quaternion-valued memristive neural networks (FQMNNs) with impulsive effect is studied here. A new mathematical expression of the quaternion-value memductance (memristance) is proposed according to the feature of the quaternion-valued memristive and a new class of FQMNNs is designed. In quaternion field, by using the framework of Filippov solutions as well as differential inclusion theoretical analysis, suitable Lyapunov-functional and some fractional inequality techniques, the existence of unique equilibrium point and Mittag-Leffler stability in finite time analysis for considered impulsive FQMNNs have been established with the order 0 < β < 1 . Then, for the fractional order β satisfying 1 < β < 2 and by ignoring the impulsive effects, a new sufficient criterion are given to ensure the finite time stability of considered new FQMNNs system by the employment of Laplace transform, Mittag-Leffler function and generalized Gronwall inequality. Furthermore, the asymptotic stability of such system with order 1 < β < 2 have been investigated. Ultimately, the accuracy and validity of obtained finite time stability criteria are supported by two numerical examples. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index