摘要
用RP/RT集成制造系统以往生产过程的工程记录数据作为样本对来训练一个改进的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RP/RT集成制造系统的三维CAD输入数据,能有效地降低集成制造系统的误差,是提高RP/RT集成制造精度的一条新途径。
Taking the production record data of RP/RT integrated manufacturing system as samples to train an improved BP nerve net, inpuring the ideal sizes of the parts which wiU be processed and taking the outlet data as the three-dimension data of the RP/RT integrated manufacturing system will greatly reduce the production errors and has been proved to be a new and effective method.
出处
《现代制造工程》
CSCD
2006年第9期26-27,66,共3页
Modern Manufacturing Engineering
基金
河南省科技攻关项目资助(001110108)
