并联坐标测量机神经网络滑模控制

Adaptive Neural Network Sliding Mode Control Method of a Parallel Coordinate Measuring Machine

研究了6-TPS型并联坐标测量机的基于RBF神经网络的直接自适应滑模控制方法。根据测量机的系统动力学模型特点,基于Lyapunov函数的综合设计方法和滑模控制理论,利用RBF神经网络与自适应技术相结合,设计了一种控制律,然后利用MATLAB进行了系统控制仿真。结果表明,测量机在有周期干扰的情况下,采用这种直接自适应神经网络滑模控制方法达到了较高的控制精度,其闭环系统具有较强的自适应性和鲁棒稳定性。

A direct adaptive neural netword （NN） sliding mode control method for 6-TPS parallel CMM （Coordinate Measuring Machine） was researched. According to its system dynamics model character, and based on of Lyapunov＇ s stability theory and method, combined RBF NN with adaptive technology, a system controller was designed, and control system numerical simulation was completed using MATLAB software platform. The simulation results show that in case of periodic interferential signal, the 6-TPS parallel CMM attains higher tracking precision using the adaptive NN sliding mode control method, its closed-loop system reveals great adapta- bility and robust stability.