| 基于BP神经网络的刀具磨损监测 |
| |
| 引用本文： | 袁琪.基于BP神经网络的刀具磨损监测[J].孝感职业技术学院学报,2010(3):99-102. |
| |
| 作者姓名： | 袁琪 |
| |
| 作者单位： | 湖北职业技术学院机电工程学院,湖北孝感432000 |
| |
| 基金项目： | 辽宁省自然科学基金项目“安装边零件数控加工中弹性变形与刀具磨损的研究”（项目编号：2005400603）的阶段性研究成果 |
| |
| 摘 要： | 结合调研资料和现场实验,文章确定了刀具磨损是影响环型零件加工精度的主要因素。针对刀具磨损,采用声发射法进行监控,运用移频小波包算法对实验采集的AE信号进行特征值提取。通过采用BP神经网络,独创性地提出将铣削时间和AE信号特征值有机结合从而建立刀具磨损监测预报模型,其模型的准确率可达96.2%,能够实时、快速、可靠地监测刀具的磨损情况。
|
| 关 键 词： | 刀具磨损 声发射 小波包 移频 BP神经网络 |
Tool Wear Monitoring Based on BP Neural Network |
| |
| YUAN Qi.Tool Wear Monitoring Based on BP Neural Network[J].Journal of Xiaogan Vocational-Technical College,2010(3):99-102. |
| |
| Authors: | YUAN Qi |
| |
| Institution: | YUAN Qi(Mechanical and Electrical Engineering School,Hubei Polytechnic Institute,Xiaogan,Hubei 432000,China) |
| |
| Abstract: | Combining the research data with field experiments,this paper indicates that the tool wear is the main factor which affects the precision of the circle part machining.In view of the tool wear,the acoustic emission is used to monitor and the features of AE signals is extracted from the experiment by means of the algorithm of wave packet frequency-shift.By using the BP neural network,the article creatively puts forward that the tool wear monitoring and forcasting model can be establised on the basis of combining the original milling time with the organic integration of the characteristics of AE signal.The accuracy of the model is 96.2% and it can be real-time,fast and reliable monitoring of tool wear. |
| |
| Keywords: | tool wear acoustic emission wavelet packet frequency-shift BP neural network |
| 本文献已被 维普 万方数据 等数据库收录！ |
