LCD降噪和全矢互信息法在不同工况下的齿轮故障诊断中的应用

发布时间：2018-03-23 04:14

  本文选题：不同工况　切入点：降噪　出处：《太阳能学报》2017年09期 　论文类型：期刊论文

【摘要】：针对不同恒定工况下的齿轮微弱故障难于诊断的问题,提出一种局部特征尺度分解(local characteristics-cale decomposition,LCD)结合全矢互信息的故障诊断方法。采用LCD对不同工况的振动信号进行分解,获取瞬时频率具有物理意义的各阶内禀尺度分量(intrinsic scale component,ISC),可消除工况所引起的频率调制及模态混叠效应所造成的干扰,再以ISC与原信号的互相关系数最大为准则进一步实现降噪。提取不同工况下的样本信号与降噪后ISC的全矢互信息绝对值之和作为样本特征向量,使用支持向量机进行分类。通过对不同工况的100组信号的识别,表明该方法能有效区分不同工况下的齿轮微弱故障特征,同时减少对人的主观经验的依赖。
[Abstract]:In order to solve the problem that it is difficult to diagnose the weak faults of gears under different constant working conditions, a local characteristic scale decomposition (characteristics-cale decompositionLCD) method combined with full vector mutual information is proposed. The vibration signals under different working conditions are decomposed by LCD. The intrinsic scale component of the intrinsic scale components of each order of the instantaneous frequency with physical significance can eliminate the interference caused by frequency modulation and modal aliasing. Then the maximum correlation number between the ISC and the original signal is taken as the criterion for further noise reduction. The sum of the absolute values of the sample signal and the total vector mutual information of the de-noised ISC under different working conditions is extracted as the sample feature vector. By using support vector machine (SVM) to classify 100 sets of signals under different working conditions, it is shown that this method can effectively distinguish the weak fault characteristics of gears under different working conditions and reduce the dependence on human subjective experience.
【作者单位】： 新疆大学机械工程学院;西安交通大学机械工程学院;
【基金】：国家自然科学基金(51565055) 新疆维吾尔自治区研究生科研创新项目(XJGRI2014025)
【分类号】：TH132.41
，

本文编号：1651842