DYNAMICAL ANALYSIS OF A SELF-CONNECTION FRACTIONAL-ORDER NEURAL NETWORK.

This paper researches the problem of bifurcation for a ring fractional-order neural network (FONN) with self-connection delay and communication delay. Self-connection delay is firstly regarded as a bifurcation parameter to examine the bifurcations of the developed FONN, and the critical values of bifurcations with respect to self-connection delay are derived. Second, communication delay is further taken as a bifurcation parameter to investigate the bifurcations of the designed FONN, and the stability zones and bifurcation points are determined. It reveals that FONN exhibits excellent stability performance when electing a lesser value of them, and the stability performance is devastated and Hopf bifurcation arises upon taking a larger one. Eventually, the efficiency of the developed theory is appraised on account of numerical verifications. [ABSTRACT FROM AUTHOR]