首页|基于自适应Wiener去噪与优化匹配追踪算法的微弱故障诊断方法

基于自适应Wiener去噪与优化匹配追踪算法的微弱故障诊断方法

扫码查看
为了准确地诊断出旋转机械的早期微弱故障,在分析信号特点的基础上,采用自适应Wiener滤波去除信号中的随机冲击干扰,并提出了一种基于优化匹配追踪算法的微弱故障特征提取方法.首先,分析了旋转机械早期微弱故障信号特点,在短时Wiener滤波中引入了窗长自适应策略,去除了早期微弱故障信号的随机冲击干扰,保留了故障信号中的有用分量;然后,分析了当前匹配追踪算法中算法迭代门限值在重构精度中存在的不足,设计了基于相邻残差的匹配追踪算法鲁棒终止条件,有效提高了故障信号的重构精度和重构信号质量;最后,将优化的匹配追踪算法应用于微弱故障特征提取中,实现了对微弱信号中强特征信号的提取目的.研究结果表明:采用自适应Wiener滤波可以有效去除信号中的随机干扰,且保留信号中的周期性故障信号;优化匹配追踪算法重构信号包络谱中的特征频率160 Hz及其倍频凸显,在其他频段的信号能量几乎为0,这意味着该方法能够准确判断出故障类型.与传统Wiener去噪和匹配追踪算法相比,自适应Wiener去噪和优化匹配追踪算法在微弱故障提取中具有可行性和优越性.
Weak fault diagnosis based on adaptive Wiener denoising and optimized matching pursuit algorithm
In order to accurately diagnose the early weak fault of rotating machinery,based on the analysis of signal characteristics,adaptive Wiener filtering was used to remove the random impact interference in the signal,and a method for extracting weak fault characteristics based on optimized matching pursuit algorithm was proposed.Firstly,the characteristics of early weak fault signals in rotating machinery were analyzed,a window length adaptive strategy was introduced in short-term Wiener filtering to achieve the removal of random impact interference,and the removal of random impact interference of early weak fault signal was realized,therefore the useful component in the fault signal was retained.Secondly,by analyzing the shortcomings of the iterative threshold in the reconstruction accuracy of the current matching pursuit algorithm,a robust termination condition based on adjacent residuals of matching pursuit algorithm was designed,and the reconstruction accuracy and quality of fault signal were effectively improved by the proposing method.Finally,the optimized matching pursuit algorithm was applied to extract the fault feature,and strong characteristic signal extraction from weak signal was realized.The simulation and experimental results show that the adaptive Wiener filter can effectively remove random interference while preserving the periodic fault signals.The characteristic frequencies 160 Hz and their harmonics in the signal envelope spectrum reconstructed by the optimized matching pursuit algorithm is significantly highlighted.In other frequency bands,the signal energy is almost zero,it means the method can accurately determine the type of fault.The results show that comparing with the traditional Wiener denoising and matching pursuit algorithm,the adaptive Wiener denoising and optimized matching pursuit algorithms have feasibility and superiority in weak fault extraction.

rotary machineadaptive Wiener denoisingmatching pursuit algorithmweak signal fault diagnosisrobust termination conditionsignal envelope

王红玉、卜令瑞、邢海燕

展开 >

山东劳动职业技术学院 信息工程系,山东 济南 271000

旋转机械 自适应Wiener去噪 匹配追踪算法 微弱信号故障诊断 鲁棒终止条件 信号包络

2024

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(12)