基于神经网络的周期扰动非线性系统自适应渐近跟踪控制  

Neural-networks-based adaptive asymptotic tracking control for nonlinear systems with periodic disturbances

作　　者：吴健 孙永波 赵前进 WU Jian;SUN Yong-bo;ZHAO Qian-jin(School of Computer and Information,Anqing Normal University,Anqing 246013,China;College of Computer Science and Engineering,Anhui University of Science&Technology,Huainan 232001,China;College of Mathematics and Big Data,Anhui University of Science&Technology,Huainan 232001,China)

机构地区：[1]安庆师范大学计算机与信息学院,安徽安庆246013 [2]安徽理工大学计算机科学与工程学院,安徽淮南232001 [3]安徽理工大学数学与大数据学院,安徽淮南232001

出　　处：《控制与决策》2022年第4期922-932,共11页Control and Decision

基　　金：国家自然科学基金项目(61603003);安徽省高校学科(专业)拔尖人才学术资助项目(gxbjZD21)。

摘　　要：针对一类具有周期扰动和输入时滞的不确定非线性系统,提出一种基于神经网络的自适应动态面控制方案.将径向基函数神经网络和傅里叶级数展开结合,构造一种混合函数逼近器来逼近系统中未知的周期扰动函数.通过引入一个积分项解决输入时滞问题,同时采用带有非线性滤波器的动态面控制方法,避免自适应反推控制方法中普遍存在的复杂性爆炸问题.所提出的神经网络控制方案能够确保闭环系统中所有信号是半全局有界的,并且系统输出能渐近地跟踪给定的参考信号.两个仿真结果验证了所提出的控制方案是有效的.A dynamic surface control scheme based on neural networks is proposed for a class of uncertain nonlinear systems with periodic disturbances and input delay.Combining the radial basis function neural network(RBFNN)with the Fourier series expansion(FSE),a mixed function approximator is constructed to approximate the unknown periodic disturbances functions in the system.An integral term is introduced to solve the problem of input delay.At the same time,the dynamic surface control method with the nonlinear filter is developed to avoid the problem of explosion of complexity commonly existed in the adaptive backstepping control method.The semiglobal boundedness of all closed-loop signals is guaranteed,and the output of the system can track a given reference signal asymptotically.Finally,two simulation results show that the proposed control scheme is effective.

关 键 词：不确定非线性系统 傅里叶级数展开 神经网络 动态面控制 周期扰动 输入时滞 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]