| 基于人工神经网络的零件特征加工方法获取及其优先权选择 |
| |
| 引用本文： | 范孝良,花广如,周雄辉. 基于人工神经网络的零件特征加工方法获取及其优先权选择[J]. 计算机集成制造系统, 2008, 14(7) |
| |
| 作者姓名： | 范孝良 花广如 周雄辉 |
| |
| 作者单位： | 华北电力大学,机械工程系,河北,保定,071003;上海交通大学,国家模具CAD工程研究中心,上海,200030 |
| |
| 基金项目： | 教育部博士点专项科研基金部分资助 |
| |
| 摘 要： | 为了获取零件特征的所有可行加工方法,并定量化描述其选择优先权,提出了一种基于人工神经网络的决策方法.在该决策方法中,基于对问题特性的分析,采用均匀设计法选取有代表性的样本集合;为避免常规反向传播算法训练网络时易陷入局部极小值的缺陷,采用可同时调整动量因子和学习率的改进反向传播算法,并用线性回归分析方法加以测试.实例验证了该方法可以有效地获取零件特征的所有可行加工方法及其选择优先权.
|
| 关 键 词： | 加工方法 优先权选择 均匀设计法 人工神经网络 反向传播 线性回归分析 |
Approach to obtain machining processes and their priority selection for part features based on ANN |
| |
| FAN Xiao-liang,HUA Guang-ru,ZHOU Xiong-hui. Approach to obtain machining processes and their priority selection for part features based on ANN[J]. Computer Integrated Manufacturing Systems, 2008, 14(7) |
| |
| Authors: | FAN Xiao-liang HUA Guang-ru ZHOU Xiong-hui |
| |
| Affiliation: | FAN Xiao-liang1,HUA Guang-ru1,ZHOU Xiong-hui2(1.Department of Mechanical Engineering,North China Electric Power University,Baoding 071003,China,2.National Die & Mold CAD Engineering Research Center,Shanghai Jiaotong University,Shanghai 200030,China) |
| |
| Abstract: | To obtain all feasible machining methods and their selection priorities described quantatively for a part feature,a decision approach based on Artificial Neural Network(ANN) was proposed.Based on the analysis of the problem characteristics,the uniform design method was adopted to select representative sample sets.To avoid the shortcoming of the general Back Propagation(BP) algorithm trapped in local minimum,the improved BP algorithm which could adjust momentum factor and learning rate simultaneously was use... |
| |
| Keywords: | machining process priority selection uniform design method artificial neural network back propagation linear regression analysis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
