基于一种新的时频分布的机械故障诊断  

Mechanical Faults Diagnosis Based on a New Time-frequency Distribution

作　　者：马瑞恒[1] 王新晴[1] 闫嘉昕[1] 王洋[1] 

机构地区：[1]解放军理工大学工程兵工程学院,江苏南京210007

出　　处：《解放军理工大学学报（自然科学版）》2003年第3期59-62,共4页Journal of PLA University of Science and Technology(Natural Science Edition)

摘　　要：在巧妙构思核函数的基础上 ,给出了一新的时频分布 TFD( Time- Frequency Distribution)及其离散算法 ,并应用于机构故障诊断。结合两个诊断实例 ,同时与 WVD( Winger- Ville Distribution)进行对比 ,发现该分布具有良好的时频聚集性 ,并且能够有效地抑制交叉项 ,表明该分布能够刻画出幅值谱和 WVD所不能反映的故障特征信号 ,能够较好地进行故障诊断。Based on the design of a new kernel function, a new Time-Frequency Distribution (TFD) is proposed and is applied to mechanical faults diagnosis. Through the diagnose is of two examples and contrast with Winger-Ville Distribution (WVD), the new distribution is found to have a high resolution and can suppress cross term effectively. In practice, this new distribution can reflect the character of the mechanical faults, which the WVD and the time-domain signal cannot, and a good method for faults diagnosis is presented.

关 键 词：核函数 时频分布 故障诊断 

分 类 号：TN911.7[电子电信—通信与信息系统] TH17[电子电信—信息与通信工程]