离心式化工泵轴承振动故障快速诊断方法研究  

Research on Rapid Diagnosis Method for Vibration Fault of Bearing in Centrifugal Chemical Pump

作　　者：翟法军 张元元 ZHAI Fajun;ZHANG Yuanyuan(Henan Xinlianxin Chemical Industry Group Co.,Ltd.,Xinxiang 453700,China)

机构地区：[1]河南心连心化学工业集团股份有限公司,河南新乡453700

出　　处：《机械与电子》2025年第2期40-44,共5页Machinery & Electronics

基　　金：河南省重点科技攻关项目(182102311012)。

摘　　要：为了确保离心式化工泵轴承的正常运行,提出一种离心式化工泵轴承振动故障快速诊断方法。结合变分模态分解与奇异值分解2种技术获取轴承振动信号的时频特征,整合轴承振动信号的时域和频域特征,构建一个多域特征集,从而实现对信号特征的全面捕捉和分析。选取不同规格和工况下已知标签数据集作为多源域,其他规格和工况下已知标签数据作为目标域,将多源域知识迁移到VGG-16深度网络,获取多个目标域模型后,将提取的具有代表性的故障特征依次输入到相同的极限学习机中展开特征融合,根据极限学习机输出分类结果,构建离心式化工泵轴承振动故障快速诊断模型,实现故障快速诊断。实验结果表明,所提方法在轴承振动故障快速诊断方面具有较高的准确率。In order to ensure the normal operation of centrifugal chemical pump bearings,a rapid diagnosis method for vibration faults of centrifugal chemical pump bearings is proposed.By combining variational mode decomposition and singular value decomposition techniques to obtain the time-frequency characteristics of bearing vibration signals,integrating the time-domain and frequency-domain characteristics of bearing vibration signals,a multi-domain feature set is constructed to achieve comprehensive capture and analysis of signal features.Selecting known label datasets under different specifications and operating conditions as the multi-source domain,and known label data under other specifications and operating conditions as the target domain,the multi-source domain knowledge is transferred to the VGG-16 deep network.After obtaining multiple target domain models,representative fault features are extracted and sequentially input into the same extreme learning machine for feature fusion.Based on the classification results output by the extreme learning machine,a fast diagnosis model for centrifugal chemical pump bearing vibration faults is constructed to achieve fast fault diagnosis.The experimental results show that the proposed method has high accuracy in the rapid diagnosis of bearing vibration faults.

关 键 词：离心式 化工泵轴承 振动故障 故障诊断 VGG-16深度网络 

分 类 号：TH133.3[机械工程—机械制造及自动化]