|
|||||
|
|
| 数控切削加工表面粗糙度RBF神经网络预测模型 | |
| 引用本文： | 曾谊晖,龚金科,李红梅,杨贤平.数控切削加工表面粗糙度RBF神经网络预测模型[J].煤矿机械,2011,32(3):117-119. |
| 作者姓名： | 曾谊晖 龚金科 李红梅 杨贤平 |
| 作者单位： | 1. 中南大学,机电工程学院,长沙,410083;湖南涉外经济学院,数控中心,长沙,410205 2. 湖南大学,机械与运载工程学院,长沙,410082 3. 湖南涉外经济学院,数控中心,长沙,410205 |
| 基金项目： | 湖南省教育厅大学生研究性学习和创新性实验计划项目，湖南省教育厅优秀青年项目 |
| 摘 要： | 针对数控切削加工表面粗糙度存在预测精度不高的问题,采用径向基(RBF)神经网络技术,以多组实际加工试验数据作为样本,建立了以转速n、进给速度vf、背吃力量ap为自变量的切削表面粗糙度预测模型。试验及预测结果表明:切削表面粗糙度RBF神经网络预测模型的预测相对误差小于2.7%,而回归分析预测值的相对误差在7.1%～14.0%变动。充分说明数控切削加工表面粗糙度RBF神经网络预测模型的预测精度高,可满足数控切削加工表面粗糙度实时在线预测的要求。 |
| 关 键 词： | 数控加工 表面粗糙度 人工神经网络 RBF算法 |
RBF Neural Network Forecasting Model of Surface Roughness for NC Machining |
|
| ZENG Yi-hui,GONG Jin-ke,LI Hong-mei,YANG Xian-ping.RBF Neural Network Forecasting Model of Surface Roughness for NC Machining[J].Coal Mine Machinery,2011,32(3):117-119. | |
| Authors: | ZENG Yi-hui GONG Jin-ke LI Hong-mei YANG Xian-ping |
| Affiliation: | 1.College of Mechanical and Electrical Engineering of Central South University,Changsha 410083,China;2.Numberical Control Tooling Center,Hunan International Economics University,Changsha 410205,China;3.College of Mechanical and Vehicle Engineering,Hunan University,Changsha 410082,China) |
| Abstract: | According to the forecasting inaccuracy of surface roughness for NC machining,using RBF neural network,a forecasting model of surface roughness is built using rotate speed n,feed peed vf,and depth of cutting ap as independent parameters.Multiunit practical machining experiments results are used as samples.The experimental and predicted results indicate that the relative error of REF networking forecasting model is less than 2.7%,the relative error of regression analysis is between 7.1%~14%.The results show that the RBF neural network forecasting model of surface roughness for NC machining has high forecasting accurate,is able to fulfill the online prediction of surface roughness for NC machining. |
| Keywords: | NC machining surface roughness artificial neural network RBF algorithm |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |