大射电望远镜柔性Stewart平台的动力学分析与控制

根据KED方法建立了柔性支腿Stewart平台的动力学模型。针对该机构为一非线性、强耦合、多输入多输出等特点,提出一种干扰观测器和基于优化神经网络结构的PID控制器(PID-NNC)相结合的控制算法来实现大射电望远镜(LT)馈源高精度轨迹跟踪。这种新方法通过构造干扰观测器来观测柔性Stewart平台的各种干扰,并对观测到的干扰信息进行补偿以抑制干扰对系统的影响,同时采用PID-NNC进行动态解耦控制。理论分析和仿真实验结果表明,该控制算法不仅能满足馈源轨迹跟踪高精度要求,而且具有较强的鲁棒性。

Dynamic Analysis and Control of the Flexible Stewart Platform for a Large Spherical Radio Telescope

The dynamic model of the flexible Stewart platform for a large spherical radio telescope(LT) was established according to the kineto elastio dynamic(KED) analysis method.Considering that the LT is non-linear and has a strong coupling,multi-input and multi-output system,a new control method that combines a disturbance observer with the PID-based improved neural network controller(PID-NNC) is utilized to realize the high-precision trajectory tracking of the feed of the LT.Through observing a variety of disturbances by using the disturbance observer,the disturbances acting on the flexible Steward platform is compensated,and the influence of the disturbances is suppressed. An algorithm for PID-NNC decoupling control is introduced to approximate optimal control.The theoretical analysis of and simulation experiments on the flexible Stewart platform indicate that the phatform has higher tracking accuracy and stronger robustness.