论文部分内容阅读
针对非线性挠性结构的模态截断 ,模型不精确与不确定性 ,以及非线性特性 ,采用了神经网络与变结构相结合的控制方法 ,针对线性化部分设计变结构控制器 ,把模型不确定性以及非线性部分作为干扰 ,用神经网络辨识切换函数的实际值与理想值之差 ,该差值反映了模型与实际系统的差别。该方法对摄动具有很强的鲁棒性 ,大大减小了系统在切换面上的抖动。针对挠性结构的模态不可测性 ,以及传感器的失误概率随其数量的增大而迅速增大的特性 ,采用了神经网络观测器观测系统状态。
Aiming at the modal truncation, model inaccuracies and uncertainties of nonlinear flexible structures, and the nonlinear characteristics, a control method based on neural networks and variable structure is adopted. The variable structure controller is designed for the linearized parts, Deterministic and non-linear parts as interference, the neural network is used to identify the difference between the actual value and the ideal value of the switching function, and the difference reflects the difference between the model and the actual system. The method is robust to perturbation and greatly reduces the jitter of the system on the switching surface. In view of the unmeasurable modal nature of flexible structures and the rapid increase of sensor error probability as the number of sensors increases, a neural network observer is used to observe the system status.