In this paper, the exponential synchronization problem for memristive delayed neural networks (MDNNs) subject to parameters mismatch is investigated. The designed controller includes both linear diffusive term and discontinuous sign function term. To further save system resources, the event-based impulsive control strategy is applied by combining impulsive control and event-triggered mechanism, while static and dynamic event-triggered conditions are proposed under this control scheme, respectively. In this case, Zeno-behavior can be excluded. Some algebraic criteria to assure exponential synchronizing of drive-response MDNNs are given by constructing suitable Lyapunov functional and employing some inequality techniques. These synchronization criteria are dependent on switching jumps, which can be chosen as small values to reduce the chattering derived from sign function term in controller. Finally, numerical simulations are given to confirm the availability of the presented results.

本文研究了参数不匹配的忆阻时延神经网络的指数同步问题。设计的控制器包括线性扩散项和不连续符号函数项。为了进一步节省系统资源，结合了脉冲控制和事件触发机制，应用了基于事件的脉冲控制策略，并在该控制方案下分别提出了静态和动态事件触发条件。在这种情况下，可以排除芝诺行为。通过构造合适的Lyapunov函数并采用一些不等式技术，给出了一些确保驱动响应MDNN指数同步的代数准则。这些同步条件取决于切换跳转，可以将其选择为较小的值，以减少从控制器中的符号函数项导出的颤动。最后，给出了数值模拟，以确认所提供结果的可用性。