| 引用本文: | 陈维,王耀南,许海霞.一类不确定系统的神经网络L2--增益鲁棒控制[J].控制理论与应用,2010,27(6):745~752.[点击复制] |
| CHEN Wei,WANG Yao-nan,XU Hai-xia.The neural network L-two-gain robust control for a class of uncertain systems[J].Control Theory and Technology,2010,27(6):745~752.[点击复制] |
|
| 一类不确定系统的神经网络L2--增益鲁棒控制 |
| The neural network L-two-gain robust control for a class of uncertain systems |
| 摘要点击 1848 全文点击 1258 投稿时间：2009-01-05 修订日期：2009-09-10 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 10.7641/j.issn.1000-8152.2010.6.CCTA090010 |
| 2010,27(6):745-752 |
| 中文关键词 L2--增益鲁棒控制 非线性系统 反步法 动态面控制方法 神经网络 |
| 英文关键词 L-two-gain robust control nonlinear systems backstepping dynamic surface control technique neural networks |
| 基金项目 国家自然科学基金重点资助项目(60835004); 湖南省教育厅资助项目(08C871). |
|
| 中文摘要 |
| 对于一类具有三角结构的单输入单输出的不确定非线性系统, 用反步法(backstepping)和动态面控制方法(dynamic surface control technique)设计了一种使用神经网络补偿未知非线性的L2--增益鲁棒控制器. 控制器设计中没有直接解HJI(Hamilton-Jacobi-Isaac)不等式. 合理的选择了L2--增益性能指标, 将被控系统各个状态变量的跟踪误差和神经网络各权值的跟踪误差看作整个控制系统的各个状态变量, 并用Lyapunov定理和HJI不等式证明了使用提出的控制器后, 这些状态变量具有小于等于事先规定的正实数γ的L2--增益. 当系统的扰动信号为零向量时, 提出的控制器在原点是大范围渐近稳定的. 仿真研究结果表明所提出的控制器具有很好的跟踪性能和很强的鲁棒性. |
| 英文摘要 |
| For a class of uncertain SISO(single-input-single-output) nonlinear systems with triangle structure, the neural network L-two-gain robust controllers are designed using backstepping and dynamic surface control technique, instead of solving the HJI(Hamilton-Jacobi-Isaac) inequality directly. A reasonable performance indicator of L-two-gain is chosen. The tracking errors of the states of the controlled system and the weights of the neural network are taken as the state variables of the whole control system. The Lyapunov theory and HJI inequality are adopted to prove that the control system has the L-two-gain which is less than or equal to the prescribed positive number γ. When the disturbance signal is a zero vector, the proposed controllers are proved to be large-scale asymptotically stable at the origin. The simulation results indicate that the proposed approach has desirable tracking performance and strong robustness. |