应用BP神经网络模型实现内圆磨床主要零件的快速反应设计

用数字化模型代替三维实体模型，将内圆磨床主要大件的BP神经网络模型放入数据库信息管理系统，应用较简单的使用方法实现了零件结构的再修改、灵敏度分析及优化．根据所提出的方法，通过从数据库信息管理系统中调用桥板的数据文件，对其进行结构参数修改、灵敏度分析及以第1阶扭转频率最高为目标的设计变量自动搜索寻优计算，结果表明将BP神经网络模型与数据库信息管理系统结合是实现零件快速反应设计的有效途径．

Rapid response design of the main parts of internal grinder based on BP neural network model

The structure updating, sensitivity analysis and optimization of the large parts of internal grinder can be carried out by means of a simple method, while the solid parts model are replaced by the digital models and the BP neural network models are loaded into the database information management system. Based on the method proposed, the structure updating, sensitivity analysis and automatic searching of the optimal solution setting the first frequency to be the largest are carried out by loading out the data files of bridge plate from the database information system. The result shows that combining BP neural network models with the database information management is an efficient way to implement the rapid response design of the parts.