基于非光滑神经网络的X-Y定位平台系统模型辨识

Modeling of X-Y positioning stage based on nonsmooth neural network

本文首先对X-Y定位平台系统的组成及模型的建立作了简单描述,接着,针对X-Y定位平台中存在的载荷以及摩擦参数的不确定性问题,提出了一种基于改进神经网络结构来对X-Y定位平台不确定性非线性系统的辨识的方法。采用非光滑神经网络来对不确定非线性系统建立模型。结果表明该方法能够对复杂的非线性系统进行辨识,比一般的神经网络具有较高的辨识精度,且具有良好的泛化性能.这种方法可以应用到工业过程实际系统中。

The X-Y table experiment system and its modeling were described firstly, following it, An improved networks was proposed to deal with the uncertainty of load and friction parameters existed in the X-Y positioning table. An nonsmooth neural-network was application to the uncertainty of nonlinear systems. Both simulations and experimental results demonstrate that the method can be identifying for the complex nonlinear systems and much improvement of performance respect to conventional neural-network, the obtained model shows good performance in model validation. The proposed approach can be applied in industries.