通过分析磨削加工过程中的误差产生的原因,使用学习训练后的BP神经网络建立了双面磨床磨削加工误差补偿模型,详细阐述了网络模型的组织结构与核心算法,提出了实现磨削加工实时误差补偿技术的硬件组成方法,并通过实验验证了模型的使用效果,结果证明,基于BP神经网络的磨削加工误差补偿模型可以有效减小磨削误差,提高磨削加工精度。 By analyzing the reason of the error in the grinding process, the BP neural network after learning training was used to establish the error compensation model of the grinding process of the double-side grinding machine. The structure and the core algorithm of the network model were elaborated. The real-time error compensation technology is used to process hardware components. The experimental results show that the BP neural network model can effectively reduce the grinding error and improve the grinding accuracy.