时频图像Hough变换在滚动轴承故障诊断中的应用  

Rolling Bearing Fault Diagnosis Using Hough Transform of Time-Frequency Image

作　　者：李宏坤[1] 张志新[1] 郭正刚[1] 邹拴庆[1] 王奉涛[1] 

机构地区：[1]大连理工大学精密与特种加工教育部重点实验室,大连116023

出　　处：《振动．测试与诊断》2010年第6期634-637,共4页Journal of Vibration,Measurement & Diagnosis

基　　金：国家自然科学基金资助项目(编号:50805014);沈鼓-大工核泵基金资助项目

摘　　要：为提高时频分析技术在设备故障诊断中的实用性,研究了时频图像处理的滚动轴承故障诊断技术。采用Hough变换对表征信号的Wigner-Ville时频分布进行分析,以提高时频谱表征非平稳信号的可靠性,特别对线性调频信号分析具有较好的实用性。以实验室的滚动轴承在不同状态下的状态识别为例,验证了时频图像Hough变换准确识别滚动轴承故障的有效性。This paper makes an investigation into the rolling bearing fault diagnosis by using time-frequency image processing technology.The Hough transform was adopted to analyze the Wigner-Ville time-frequency distribution to improve the accuracy of nonstationary signal analysis,especially for the linear frequency modulation signal investigation.Pattern recognition of rolling bearing conditions was studied.The results show that the Hough transform of time-frequency image can effectively classify the rolling bearing faults.

关 键 词：时频分析 HOUGH变换 WIGNER-VILLE分布 故障诊断 

分 类 号：TN911[电子电信—通信与信息系统] TH113[电子电信—信息与通信工程]