基于MMED+TQWT方法的叶轮振动信号特征提取研究
Research on Bearing Fault Feature Extraction and Diagnosis Based on MMED+TQWT Method
袁艳 1辛保娟1
作者信息
- 1. 西安交通工程学院 机械与电气工程学院,陕西 西安 710000
- 折叠
摘要
为了提高高炉煤气余压发电装置(TRT)叶轮故障诊断能力,开发了以最小熵解卷积MMED和可调品质因子小波变换TQWT两种方法共同诊断叶轮故障的技术.先利用MMED方法转换初始振动信号获得更明显的故障冲击成分,再对经过预处理的信号实施TQWT分解,并设定相应的品质因子Q,再按照峭度最大原则确定子带最优分量并计算包络谱数据,实现叶轮故障的诊断功能.研究结果表明:采用本文方法分析故障冲击成分获得了显著增强,对噪声干扰起到了明显抑制作用.从包络谱内明显看到跟叶轮故障特征频率相同的频率特征,形成了明显的边频带,说明叶轮中已形成故障特征.
Abstract
In order to improve the fault diagnosis ability of blast-furnace gas residual pressure power plant(TRT)impeller,two methods of minimum entropy deconvolution MMED and adjustable quality factor wavelet transform TQWT are developed to diagnose impeller fault.First,MMED method is used to convert the initial vibra-tion signal to obtain more obvious fault impact components,and then TQWT decomposition is implemented on the pre-processed signals,and the corresponding quality factor Q is set,and then the optimal subband component is determined according to the kurtosis maximum principle and the envelope spectrum data is calculated to realize the fault diagnosis function of the impeller.The results show that the proposed method can significantly enhance the fault impact components and restrain the noise interference.From the envelope spectrum,it is obvious that the fre-quency characteristics are the same as the fault characteristic frequency of the impeller,forming a clear side fre-quency band,indicating that the fault characteristics have been formed in the impeller.
关键词
叶轮/最小熵解卷积/可调品质因子小波变换/特征提取/故障诊断Key words
bearing/minimum entropy deconvolution/adjustable quality factor wavelet transform/feature ex-traction/fault diagnosis引用本文复制引用
基金项目
西安交通工程学院中青年基金项目(2022KY-09)
出版年
2024
NETL
NSTL
万方数据