基于振动分析技术的数控机床主轴故障预测方法
Fault prediction Method of Computer Numerical Control Machine Spindle Based on Vibration Analysis Technology
刘刚1
作者信息
- 1. 安徽容知日新科技股份有限公司,合肥 230000
- 折叠
摘要
探讨基于振动分析技术的数控机床主轴故障预测方法.首先深入分析高速精密数控磨床主轴的振动特性,其次采用高灵敏度传感器采集振动信号,并运用小波阈值去噪法去除高频随机噪声和电磁干扰,再次利用短时傅里叶变换提取特征参数,最后结合支持向量机与AdaBoost算法建立故障预测模型.实验结果显示,该方法具有较高的故障早期识别率和预警提前量,计算响应时间能够满足在线监测要求.
Abstract
This paper discusses the fault prediction method of computer numerical control machine tool spindle based on vibration analysis technology.Firstly,the vibration characteristics of the spindle of high-speed precision computer numerical control grinding machine are analyzed deeply.Secondly,the vibration signals are collected by high-sensitivity sensors,and the high-frequency random noise and electromagnetic interference are removed by wavelet threshold denoising method.Then the feature parameters are extracted by short-time Fourier transform.Finally,a fault prediction model is established by combining support vector machine and AdaBoost algorithm.The experimental results show that this method has high early fault recognition rate and early warning lead,and the calculated response time can meet the requirements of online monitoring.
关键词
数控机床主轴/故障预测/振动分析技术Key words
computer numerical control machine tool spindle/failure prediction/vibration analysis technique引用本文复制引用
出版年
2024
NETL
NSTL
万方数据