基于磁悬浮作动器的自适应有源振动控制研究

针对周期扰动提出一种基于磁悬浮作动器的非线性前馈自适应有源振动控制算法。算法中将磁悬浮作动器视为具有时变非线性的单输入输出系统,并使用径向基函数神经网络进行控制,分别采用聚类算法和随机梯度算法对其隐层中心点和输出层权值进行自适应调整。该算法摆脱了传统磁悬浮控制对模型的依赖,在正常工作条件下不需对作动器建模。仿真和实验结果表明:在单自由度主动隔振系统中,非线性自适应算法可以显著降低周期振动的能量,同时能对磁悬浮作动器的时变非线性进行有效的补偿。 

Research on adaptive active vibration control using maglev actuator

A feedforward adaptive active vibration control (AVC) algorithm using maglev actuator is proposed to attenuate periodic disturbance in this paper. Maglev actuator is viewed as a single-input-single-output (SISO) system with time-varying nonlinearity in the proposed algorithm. A radial basis function (RBF) neural network is used as the controller, whose hidden-layer centers and output-layer weights are adapted according to clustering algorithms and stochastic gradient algorithms respectively. In the proposed algorithm, there is no need to model the maglev actuator under normal circumstance, which is critical and difficult in conventional maglev control. The results of simulations and experiments based on a single degree-of-freedom vibration isolation system show that the adaptive algorithm can greatly attenuate the periodic disturbance, and can efficiently compensate for the actuator's time-varying nonlinearity at the same time.