基于分形几何状态监测方法的应用研究  

RESEARCH ON CONDITION MONITORING OF MACHINERY USING FRACTAL GEOMETRY

作　　者：刘天雄[1] 华宏星[1] 李中付[1] 陈兆能[1] 朱继梅[2] 

机构地区：[1]上海交通大学机械工程学院,上海200030 [2]上海理工大学

出　　处：《机械工程学报》2001年第5期100-104,共5页Journal of Mechanical Engineering

基　　金：国家自然科学基金资助项目! (19972 0 16 )

摘　　要：根据二进小波变换原理 ,分析了机械设备噪声信号某一特征尺度下的时域信号分量的分形特性 ,提出可以用分形关联维数描述设备的工作特征 ,进而对状态进行监测。试验表明 ,相同工作状态下 ,某一特征尺度下的时域信号分量有相近的关联维数 ,而在不相同状态下则有明显不同的关联维数。关联维数可以作为机械设备状态分类的重要依据 。The wavelet transformation based on the noise signal of working machinery is established for the analysis of the correlation dimension of the typical working condition. Correlation dimension of the decomposing coefficient of wavelet transformation is calculated to recognize the working condition and detect the working condition. The field experiment show that the correlation dimension of the similar working conditions have some similar value while different working conditions have some distinct difference. The correlation plot of typical working condition of pump plotted in the same paper has different domain. The experimental results and correlation plot confirmed the feasibility and validity of the method proposed.

关 键 词：状态监测 小波变换 分形几何 关联维数 故障诊断 机械设备 

分 类 号：TH17[机械工程—机械制造及自动化]