自适应事件触发下时滞神经网络的滑模控制器设计

Design of Sliding Mode Controller for Delayed Neural Networks under Adaptive Event-Triggering

摘要: 针对一类具有时变时滞的神经网络系统，本文提出一种基于自适应事件触发机制的滑模控制策略，目的是为有效节约通信资源并抑制系统不确定性及外部干扰的影响。首先，设计一种与时变时滞的连续时间神经网络系统状态相关的自适应事件触发条件，以动态调节数据传输频率。其次，构造Lyapunov泛函，利用自由权矩阵方法处理时滞项，证明了闭环系统的渐近稳定性的充分条件并在此基础上得到了相应的滑模控制器，求解控制器增益与事件触发参数，以保证系统状态在有限时间内到达并维持在滑模面。最后，通过数值仿真验证了所提控制方案在节省通信资源的同时，能够有效镇定系统并对匹配不确定性具有强鲁棒性。

Abstract: For a class of neural network systems with time-varying delays, this paper proposes a sliding mode control strategy based on an adaptive event-triggering mechanism, aiming to effectively save communication resources and suppress the impact of system uncertainties and external disturbances. Firstly, an adaptive event-triggering condition related to the state of the continuous-time neural network system with time-varying delays is designed to dynamically adjust the data transmission frequency. Secondly, a Lyapunov functional is constructed, and the free-weighting matrix method is used to handle the delay terms. Sufficient conditions for the asymptotic stability of the closed-loop system are proved, and the corresponding sliding mode controller is derived on this basis. The controller gains and event-triggering parameters are solved to ensure that the system states reach and remain on the sliding surface within a finite time. Finally, numerical simulations verify that the proposed control scheme can effectively stabilize the system and has strong robustness against matched uncertainties while saving communication resources.