摘要
齿轮出现故障时,齿轮的故障信息包含在齿轮的振动信号中,用合适的特征提取方法提取故障信息是故障诊断关键又困难的问题。针对这一问题,提出了一种变分模态分解(VMD)与平均能量结合的齿轮故障特征提取方法。该方法首先用变分模态分解的方法将实验室采集到的各类振动信号分别进行变分模态分解,然后对分解得到的每一个模态分量求平均能量作为齿轮故障特征量。为了验证提取到的齿轮故障特征的准确性,采用欧氏距离方法对齿轮故障特征进行分类和诊断。诊断结果表明,本方法的诊断正确率达到100%。因此,所提出的基于VMD和平均能量的特征提取方法能够准确地提取到齿轮故障特征。
When the gear is faulty,the fault information is included in the gear vibration signals. Extracting fault information with appropriate feature extraction method is a key difficulty in fault diagnosis. To solve this problem,a newfault feature extraction method based on variational mode decomposition（ VMD） and average energy was proposed. In this method,the vibration signals collected from laboratory were first decomposed by VMD. Then the obtained average energy of each mode was taken as the gear fault feature. In order to verify the accuracy of the extracted gear fault feature,the Euclidean distance method was used to classify and diagnose the gear fault feature. Diagnosis results showthat the correct rate of the proposed method is100%. Thus,the proposed feature extraction method based on VMD and average energy can accurately extract gear fault feature.
作者
蒋丽英
高爽
崔建国
于明月
卢晓东
王景霖
JIANG Li-ying GAO Shuang CUI Jian-guo YU Ming-yue LU Xiao-dong WANG Jing-lin(College of Automation, Shenyang Aerospace University, Shenyang 110136, China Key Laboratory of Aviation Science and Technology on Fault Diagnosis and Health Management, AVIC Shanghai Aero Measurement ＆ Control Technology Research Institute, Shanghai 201601, China)
出处
《沈阳航空航天大学学报》
2016年第6期59-65,共7页
Journal of Shenyang Aerospace University
基金
国家自然科学基金青年基金(项目编号:51605309)
航空科学基金(项目编号20153354005)
关键词
齿轮
振动信号
变分模态分解
特征提取
gear
vibration signal
variational mode decomposition
feature extraction
