基于小波分析的高速CNC集成制造工况健康监控技术研究

发布时间：2018-02-22 03:38

本文关键词： 小波分析健康监控高速加工神经网络　出处：《河北工业大学》2011年硕士论文　论文类型：学位论文

【摘要】：随着高速加工技术的发展,制造过程中的工况监控所遇到的问题也越来越突出,且直接影响了生产效率,因此对高速CNC集成制造过程进行监控,才能保证产品质量,提高加工效率,延长刀具使用寿命,确保人员和设备的安全。把结构健康监控的概念引入到高速CNC集成制造系统中,可用一些工况信息(如切削力、扭矩、功率、主轴变速、振动、热变形等)来描述数控加工过程中的机床的各种状态。高速CNC集成制造健康监控涉及到主轴发热监测、滚珠丝杠发热监测、刀具磨损状态监测、工件加工状态监测等。本文着重研究刀具的工况健康监控方法,提出的方法也使用与其他方面。本文利用振动传感器对刀具磨损信号进行采集,并对所采集的信号进行分析处理,通过对比不同磨损程度的刀具信号,提取与刀具磨损相关的特征值。时域方面,提取出信号的均方根作为特征值;时频域方面,采用了小波包分解频带能量监测法,对振动信号进行了频段能量统计,提取出信号的特征频段能量作为特征值。再此基础上建立基于BP神经网络和D-S证据理论相结合的车刀故障综合诊断模型,丰富和发展了刀具磨损监控技术。最后在上述研究的基础上,构建了VC与MATLAB混合编程开发系统数据处理平台
[Abstract]:With the development of high speed machining technology, the problems of working condition monitoring in manufacturing process are more and more prominent, and directly affect the production efficiency. Therefore, monitoring the high speed CNC integrated manufacturing process can ensure the product quality. The concept of structural health monitoring is introduced into the high speed CNC integrated manufacturing system, and some working conditions information (such as cutting force, torque, power, spindle speed change, etc.) can be used to improve the machining efficiency, prolong the tool life and ensure the safety of the personnel and equipment. High speed CNC integrated manufacturing health monitoring involves spindle heating monitoring, ball screw heating monitoring, tool wear monitoring, This paper focuses on the research of the tool condition health monitoring method, the proposed method is also used and other aspects. In this paper, the tool wear signal is collected by vibration sensor, and the collected signal is analyzed and processed. By comparing the tool signal with different wear degree, the characteristic value related to tool wear is extracted. The root mean square (RMS) of the signal is extracted as the eigenvalue, and in time-frequency domain, the frequency band energy monitoring method based on wavelet packet decomposition is used to calculate the frequency band energy of the vibration signal. The characteristic frequency band energy of the signal is extracted as the eigenvalue, and then a comprehensive fault diagnosis model of turning tool based on BP neural network and D-S evidence theory is established. The tool wear monitoring technology is enriched and developed. Finally, the data processing platform of VC and MATLAB hybrid programming system is constructed based on the above research.
【学位授予单位】：河北工业大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH166

【参考文献】