音圈电机定位的神经网络PID前馈控制方法

Neural Network PID Feedforward Control Method for the Position of Voice Coil Motor

根据摄像头模组封装生产定位的控制特点,分析了音圈电机定位系统的数学模型,提出了音圈电机定位的神经网络PID前馈控制模型,利用BP神经网络在线自整定PID参数。为弥补BP神经网络的不足,对BP神经网络算法进行了改进。将所提出的模型用于摄像头模组校正装置上,实验和应用结果表明,所提出的控制模型具有较高的定位精度、较快的调节速度,能够满足提高生产效率的实际需要,可以应用于电子产品封装等类似工程领域。

According to the positioning control characteristics of camera module in the packaging process,the mathematical model for the positioning system of voice coil motor was analyzed,and the neural network PID feedforward control model for the position of voice coil motor was proposed,which could self-tuning PID parameters online. In order to compensate BP neural network,BP neural network algorithm was improved. By applying the proposed model to the camera module calibration device,the experiment and application results show that the model has high positioning accuracy,and fast regulating speed,which can improve the efficiency of production,and can be applied to electronic product packaging,and other areas of the similar engineering.