论文部分内容阅读
针对精密定位装置存在非线性,精确数学模型难于建立的缺陷,提出了精密定位的神经网络控制方法.将BP神经网络应用于该控制系统中,系统以光栅常数100μm的光栅为定位标记,以激光衍射产生的莫尔光光强及光强的变化率为神经网络的输入变量,利用神经网络的自学习功能进行精密定位控制.建立了精密定位的神经网络控制模型,模型由输入层、隐层和输出层3层神经元组成,通过对光强及光强变化率的映射,得到电机驱动信号.实验结果表明,使用神经网络控制,控制响应快,稳定性好,鲁棒性强,可有效改善控制质量,提高定位速度,系统可获得±0·5μm的定位精度.
Aiming at the defect that the precision positioning device has nonlinearity and the precise mathematical model is hard to establish, a precise positioning neural network control method is proposed. BP neural network is applied to this control system. The system uses the grating with a grating constant of 100μm as the positioning mark, The light intensity and the rate of change of light generated by diffraction are the input variables of neural network, and the precise positioning control is realized by using the self-learning function of neural network.The neural network control model of precision positioning is established.The model consists of input layer, hidden layer And the output layer 3-layer neurons, the motor drive signal is obtained through the mapping of the light intensity and the rate of change of light intensity.Experimental results show that using neural network control, the control response is fast, the stability is good, the robustness is good, and it is effective Improve the control quality and improve the positioning speed, the system can obtain the positioning accuracy of ± 0.5μm.