BP神经网络在彩色光固化快速成型系统误差控制中的应用  

Application of BP Neural Network in the Error Control of Color SL Rapid Prototyping System

作　　者：王鹏[1] 王向伟[1] 郑大宇[1] 

机构地区：[1]哈尔滨商业大学,哈尔滨150028

出　　处：《机械工程师》2008年第8期39-40,共2页Mechanical Engineer

基　　金：黑龙江省教育厅项目(11511107)

摘　　要：用彩色光固化快速成型系统以往生产过程的工程记录数据作为样本对,训练一个改进的BP神经网络,把要加工制件的理想尺寸输入已训练好的神经网络的输入层,以该神经网络的输出层数据作为彩色光固化快速成型制造系统的三维CAD输入数据,能有效地降低集成制造系统的误差,提高彩色光固化快速成型制造精度。Using the previous project data records of the production process in the color SL rapid prototyping system as a sample to train a neural network to improve the BP. Inputting the ideal size of the processing parts to the input layer of the neural network which has been trained, and the output layer data of the neural network is the 3D CAD data entry of color SL rapid prototyping manufacturing system. It can effectively reduce the error of integrated manufacturing system, and it is a new way of improving the color SL rapid prototyping manufacturing accuracy.

关 键 词：BP神经网络 彩色光固化快速成型系统 精度 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]