引用本文:游星星,陶栩,郭斌,向国菲,刘凯,佃松宜.基于观测器和事件触发的分数阶非线性系统神经网络控制[J].控制理论与应用,2024,41(10):1735~1744.[点击复制]
YOU Xing-xing,TAO Xu,GUO Bin,XIANG Guo-fei,LIU Kai,DIAN Song-yi.Neural network control for fractional-order nonlinear systems based on observer and event-triggered strategy[J].Control Theory and Technology,2024,41(10):1735~1744.[点击复制]
基于观测器和事件触发的分数阶非线性系统神经网络控制
Neural network control for fractional-order nonlinear systems based on observer and event-triggered strategy
摘要点击 3913  全文点击 127  投稿时间：2022-09-05  修订日期：2024-02-25
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2023.20782
  2024,41(10):1735-1744
中文关键词  分数阶非线性系统  观测器  神经网络控制  事件触发控制  动态面控制
英文关键词  fractional-order nonlinear systems  observer  neural networks control  event-triggered control  dynamic surface control
基金项目  中央高校基本科研业务费项目(2024SCU12080), 国家资助博士后研究人员计划项目(GZC20231783), 四川省自然科学基金项目(2021YJ0092, 2023NSFSC0475)资助.
作者单位E-mail
游星星 四川大学 youxingxing@stu.scu.edu.cn 
陶栩 四川三联新材料有限公司  
郭斌* 四川大学  
向国菲 四川大学  
刘凯 四川大学  
佃松宜 四川大学  
中文摘要
      针对一类分数阶非线性系统的跟踪控制问题, 本文提出了一种自适应神经网络事件触发控制方案. 首先,利用径向基函数神经网络来逼近未知的非线性函数, 构造了基于神经网络的状态观测器估计原系统状态. 然后, 在控制器设计中引入了事件触发策略, 通过Lyapunov方法分析了闭环系统的稳定性. 本文提出了一个新条件来估计事件触发条件的时间间隔下限, 避免了Zeno现象. 理论分析表明, 提出的控制方案不仅能确保跟踪误差收敛到原点附近的邻域内, 而且保证了闭环系统中所有信号的有界性. 最后, 分数阶互联电力系统仿真展示了方案的有效性.
英文摘要
      In this paper, an adaptive neural network event-triggered control scheme is proposed for tracking control of a class of fractional-order nonlinear systems. Firstly, radial basis function neural networks are used to approximate the unknown nonlinear functions, and a neural networks-based state observer is constructed to estimate the state of original system. Then, the event-triggered strategy is presented in the design of controller, and the stability of closed-loop system is analyzed by using the Lyapunovmethod. In addition, a new condition is developed for estimating the time interval lower bound of the event-triggered condition of fractional-order nonlinear systems in this paper, thus the Zenophenomenon can be avoided. Theoretical analysis shows that the proposed control scheme can not only ensure that the tracking error converges to the neighborhood near the origin, but also can guarantee the boundedness of all signals in the closed-loop system. Finally, simulation of fractional-order interconnected power systems demonstrates the effectiveness of scheme.