冲床保护装置的人手与坯料神经网络分类器模型

作者：孙伟东1,袁昌明2,王 强2，蒋 庆2 日期：2007-04-10/span> 浏览：4145 查看PDF文档

                                      冲床保护装置的人手与坯料神经网络分类器模型
                                       孙伟东1,袁昌明2,王 强2，蒋 庆2
(1.上海理工大学 光学与电子工程学院,上海 200093；2.中国计量学院 计量技术工程学院,浙江 杭州 310018 )

摘 要：为建立冲压作业过程中人手与坯料的神经网络分类器模型，以解决冲压时不能连续进料的问题，采用了BP算法对人手和坯料的温度数据对进行训练，并利用样本估计其错误率小于4%。实验结果表明，此BP分类器模型结构简单，错误率低，对冲压设备智能安全保护装置的研究有很好的指导作用。
关键词：冲床；BP算法;分类器模型;样本估计;错误率

          Classification model of manhand and billet based on BP algorithm in safety protector for puncher
                  SUN Weidong1, YUAN Changming2, WANG Qiang2, JIANG Qing2
(1.College of Optics and Electronics Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China;2.College of Metrological Technology Engineering, China Jiliang University, Hangzhou 310018, China)

Abstract: In order to construct a neural network classifier of manhand and billet in stamping process of factory, datasets of temperature of manhand and billet was trained with BP algorithm. Error rate was estimated with sample estimation and it was up to less than 4%. Experimental results show the classifier is simple and effectively reduce computation burden. The model can guide the study of the intelligent protection against safety accident in the process of  stamping.
Key words: puncher; back propagation algorithm; classification model; sample estimation; error recognition rate

参考文献：
［1］王亚良,计时鸣,费新章，等.基于计算机图像处理的冲压床保护技术研究［J］.机床与液压,2004,(9):60-62.
［2］刘 炎,巩水利,马艳军.高灵敏度机床安全保护系统的研制［J］.红外技术,2000，(11):45-46.
［3］晓 溪.安全型压力机在沪研制成功［J］.锻压机械,2000，(4):42.
［4］魏海坤.神经网络结构设计的理论与方法［M］.北京:国防工业出版社,2005.
［5］DUDA R O, HART P E, STORK D G. Pattern Classification: Second Edition［M］.北京:机械工业出版社,2004.
［6］KEI B. Neural networks and fuzzy system:A Dynamical Approach to Machine Intelligence［M］.NJ: PrentHall,1992.
［7］高 隽.人工神经网络原理及仿真实例［M］.北京:机械工业出版社,2003.
［8］飞思科技产品研发中心.神经网络理论与Matlab7实现［M］.北京:电子工业出版社,2005.
［9］董长虹.Matlab神经网络与应用［M］.北京:国防工业出版社,2005.
［10］边肇祺,张学工.模式识别：第二版［M］.北京:清华大学出版社,2000.