针对一类二阶时滞惯性神经网络模型,提出一种基于时滞反馈的分岔控制方法.利用时滞微分方程动力学理论,给出反馈控制系统的稳定性以及发生Hopf分岔的判别条件.数值仿真显示所设计的控制器不仅能有效延迟网络分岔的发生,还能扩大稳定域并改善网络的收敛速度. Aiming at a class of second-order inertial neural network with time-delay model, a bifurcation control method based on time-delay feedback is proposed. The stability of the feedback control system and the discriminant conditions of Hopf bifurcation are given by using the dynamical theory of delay differential equations The numerical simulation shows that the designed controller can not only effectively delay the occurrence of network bifurcation, but also expand the stable region and improve the network convergence speed.