摘要
针对往复压缩机轴承振动信号的复杂多分量耦合特性,提出了一种基于参数优化VMD和MDE的往复压缩机轴承故障诊断方法。利用遗传算法搜索VMD算法的最佳影响参数组合,确定VMD算法需要设定的带宽参数和分量个数对故障信号分解。计算分解后各BLIMF分量的峭度值,筛选出最佳BLIMF分量并重构故障信号,然后对重构后故障信号进行MDE分析形成故障特征向量,输入到极限学习机中进行分类识别。对往复压缩机轴承故障实测信号进行分析,实验结果表明,该方法可有效提取出往复压缩机轴承故障特征,特征向量具有较好可分性,实现了往复压缩机轴承故障特征的有效识别。
Aiming at the complex multi-component coupling characteristics of the reciprocating compressor bearing vibration signal, a fault diagnosis method for reciprocating compressor bearing based on parameter optimization VMD and MDE is proposed. The genetic algorithm was used to search the best influence parameter combination of the VMD algorithm, and the bandwidth parameters and component numbers that need to be set by the VMD algorithm were determined to decompose the fault signal. The kurtosis value of each BLIMF component after decomposition was calculated to select the best BLIMF component, and the fault signal was reconstructed. Then the MDE analysis was used to form the fault eigenvector and input into the extreme learning machine for classification and identification. The measured signal of the bearing fault of the reciprocating compressor is analyzed. The experimental results show that the method can effectively extract the bearing fault characteristics of reciprocating compressor, and the eigenvector has good separability, which realizes the effective identification of the bearing fault characteristics of reciprocating compressor.
作者
李颖
王金东
赵海洋
宋美萍
刘著
LI Ying;WANG Jin-dong;ZHAO Hai-yang;SONG Mei-ping;LIU Zhu(School of Mechanical Science and Engineering,Northeast Petroleum University,Daqing Heilongjiang163318,China)
出处
《组合机床与自动化加工技术》
北大核心
2019年第4期120-123,132,共5页
Modular Machine Tool & Automatic Manufacturing Technique
基金
黑龙江省自然科学基金资助项目(E2016009)
东北石油大学研究生创新科研项目资助(YJSCX2017-020NEPU)
东北石油大学青年科学基金项目(2018ANC-31)
关键词
往复压缩机
轴承
参数优化VMD
MDE
reciprocating compressor
bearing
parameter optimization VMD
MDE
