首页|基于DE-VMD和GMDE的往复压缩机轴承间隙故障诊断方法

基于DE-VMD和GMDE的往复压缩机轴承间隙故障诊断方法

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
针对往复压缩机轴承间隙故障特征提取困难、识别准确率不高等问题,提出了差分进化算法优化变分模态分解方法和广义多尺度散布熵相结合的往复压缩机间隙故障诊断方法.首先,采用差分进化算法对变分模态分解算法的两个核心参数进行了优化,并利用优化后的变分模态分解方法对轴承间隙振动信号进行了信号分解和重构处理;然后,研究了多尺度散布熵的粗粒化过程,通过将方差粗粒化代替均值粗粒化,进行了多尺度处理,构建了广义多尺度散布熵算法,利用广义多尺度散布熵算法对重构信号进行了故障特征提取分析;最后,设计了核极限学习机模型对故障特征向量集进行了分类识别,完成了往复压缩机轴承间隙不同故障状态的智能诊断研究.研究结果表明,该故障诊断方法的识别准确率高达97%,高效地实现了轴承不同种类故障的智能诊断目的.
Fault diagnosis method for bearing clearance of reciprocating compressor based on DE-VMD and GMDE
Aiming at the difficulty of feature extraction and low recognition accuracy of bearing clearance fault of reciprocating compressor,a fault diagnosis method of bearing clearance fault of reciprocating compressor was proposed by combining differential evolution(DE)algorithm optimization variational mode decomposition(VMD)method and generalized multi-scale dispersal entropy.Firstly,the differential evolution algorithm was used to optimize the two core parameters of the variational mode decomposition algorithm,and the optimized variational mode decomposition method was used to decompose and reconstruct the vibration signal of the bearing clearance.Then,the coarse-grained process of the multi-scale dispersal entropy algorithm(MDE)was studied,and the generalized multi-scale dispersal entropy algorithm(GMDE)was constructed by using variance coarse-grained instead of mean coarse-grained to carry out multi-scale processing.Finally,the kernel extreme learning machine model(KELM)was designed to classify and identify the fault feature vector set,and the intelligent diagnosis of different fault states of the bearing clearance of reciprocating compressor was completed.The research results show that the identification accuracy of this fault diagnosis method is as high as 97%,and the intelligent diagnosis of different kinds of bearing faults is realized efficiently.

reciprocating compressorbearing fault diagnosisvariational mode decomposition(VMD)generalized multi-scale dispersion entropy(GMDE)kernel extreme learning machine(KELM)differential evolution(DE)algorithm

李彦阳、蔡剑华、曲孝海

展开 >

东北石油大学 机械科学与工程学院,黑龙江 大庆 163318

黑龙江八一农垦大学 土木水利学院,黑龙江 大庆 163319

湖南文理学院 数理学院,湖南 常德 415000

往复压缩机 轴承故障诊断 变分模态分解 广义多尺度散布熵 核极限学习机 差分进化算法

黑龙江自然科学基金联合引导项目湖南文理学院科学研究项目

LH2021E02122ZD08

2024

10.3969/j.issn.1001-4551.2024.04.014

机电工程
浙江大学　浙江省机电集团有限公司

机电工程

CSTPCD北大核心
影响因子：0.785
ISSN：1001-4551
年,卷(期)：2024.41(4)
  • 18