基于分形滤波和局域波分解的齿轮箱轴承故障诊断  

Gearbox Bearings Fault Diagnosis Based on Multi-fractal Theory and Local Wave Decomposition

作　　者：潘铭志[1] 王莲花[2] 许昕[3] 潘宏侠[3] 

机构地区：[1]晋中学院机械学院,山西晋中030619 [2]包头职业技术学院,内蒙古包头014030 [3]中北大学机械与动力工程学院,山西太原030051

出　　处：《机械工程与自动化》2016年第6期136-138,共3页Mechanical Engineering & Automation

基　　金：国家自然科学基金资助项目(50875247);国家自然科学基金资助项目(51175480);山西省教育厅高等学校科技创新项目(2014148)

摘　　要：提出了一种基于分形滤波和局域波分解对齿轮箱进行故障诊断的方法。针对齿轮箱工作环境的影响,传感器采集的信号含有噪声干扰,先使用分维数作为模糊系统控制系数的分形滤波器对振动加速度信号进行处理,以提高信号信噪比,然后用局域波分解算法将消噪数据进行处理,得到基本模型分量和剩余成分,最后对每一个基本模型分量做出功率谱图进行频谱分析,从而识别出齿轮箱轴承故障。A method based on fractal filtering and local wave decomposition for gearbox fault diagnosis is presented in this paper.The gearbox will be disturbed by environmental factors during operation,so that the data collected by sensors contain unwanted noise.Firstly,a short time-division-dimensional filter using short-time fractal fuzzy control parameters is applied to preprocessing the vibration signal to improve the signal to noise ratio.Then,the signal is decomposed into two parts,which are the basic model components and the remaining component.At last,the power spectra of each basic model component is made in order to identify the gearbox failure.

关 键 词：分形滤波 局域波分解 齿轮箱轴承 故障诊断 

分 类 号：TH165.3[机械工程—机械制造及自动化]