摘要
轴承是核电厂旋转机械的重要支撑部件,为了提高轴承早期故障的检测能力,本文提出了一种基于人工蜂群优化的参数自适应变分模态分解故障特征提取方法。利用峭度和相关系数构建加权峭度指标;以最大加权峭度指标为目标函数,利用人工蜂群算法对变分模态分解过程中的模态数和带宽控制参数进行优化,获取最优参数组合并对轴承振动信号进行模态分解;对加权峭度指标最大的敏感模态分量进行包络谱分析并识别故障频率。通过仿真与实验验证了该方法的有效性,并通过与集成经验模态分解、局部均值分解和固定参数变分模态分解的特征提取效果进行比较,突出了该方法在轴承早期故障诊断中的优势。
Bearings are important supporting components of rotating machinery in nuclear power plants.To improve the early fault detection of bearings,a fault feature extraction method based on parameter-adaptive variational mode decomposition(VMD)optimized by the artificial bee colony(ABC)algorithm is proposed in this paper.First,a measurement index termed the weighted kurtosis index is constructed by the kurtosis index and correlation coefficient.Then,taking the maximum weighted kurtosis index as the objective function,the ABC algorithm is used to optimize the mode number and mode frequency bandwidth control parameter of VMD to obtain the optimal parameter combination and mode decomposition of bearing vibration signals.Finally,the envelope spectrum analysis is performed on the sensitive mode with the maximum weighted kurtosis index to identify the fault characteristic frequency.A simulation and experiment validate the effectiveness of the proposed method,and comparisons with ensemble empirical mode decomposition,local mean decomposition,and fixed-parameter VMD highlight the advantages of the proposed method in the early fault detection of bearings.
作者
朱少民
夏虹
王志超
彭彬森
姜莹莹
张汲宇
ZHU Shaomin;XIA Hong;WANG Zhichao;PENG Binsen;JIANG Yingying;ZHANG Jiyu(Key Laboratory of Nuclear Safety and Advanced Nuclear Energy Technology, Ministry of Industry and Information Technology, Harbin Engineering University, Harbin 150001, China;Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, Harbin 150001, China)
出处
《哈尔滨工程大学学报》
EI
CAS
CSCD
北大核心
2021年第10期1550-1556,共7页
Journal of Harbin Engineering University
基金
国家自然科学基金项目(51379046)
黑龙江省自然基金项目(E2017023).
关键词
轴承
变分模态分解
人工蜂群
加权峭度
包络谱分析
核电厂
信号处理
特征提取
故障诊断
bearing
variational mode decomposition(VMD)
artificial bee colony(ABC)
weighted kurtosis index
envelope spectrum analysis
nuclear power plant
signal processing
feature extraction
fault diagnosis
