柔性神经网络滑模主动控制技术

针对复杂激励条件下的振动控制,对Jiles-atherton模型的磁致伸缩作动器在双层隔振系统中的主动控制进行了研究。以传统滑模控制为基础,提出一种柔性神经网络滑模控制算法。用正则化方法设计控制器的切换矩阵,建立神经网络权值和柔性映射参数更新学习公式,并将该控制策略应用于双层隔振系统的振动主动控制中。通过单频、多频及随机信号激励进行仿真研究,结果表明:柔性神经网络滑模控制器具有较强的鲁棒性,具有较好的控制效果。

Research on Active Control using Flexible Neural Network Sliding Mode Algorithm

For the vibration control in complex excitation, the active control of double-layer vibration isolation system using magnetostrictive actuator of Jiles-atherton model is researched. Based on the traditional sliding mode control, a flexible neural network sliding mode control algorithm is proposed and the controller switching matrix is designed by the regularization method, then the updating formula of the neural network weights and flexible mapping parameter are also established. Furthermore the control strategy is used for the active vibration control in double- layer vibration isolation system. At the last the single frequency, multi frequency and random signal excitation are simulated and the results show that: the flexible neural network sliding mode controller has a strong robustness, which can make the active control significantly attenuate the disturbance in double- layer vibration isolation system.