基于时间-小波能量谱样本熵的滚动轴承智能诊断方法

发布时间：2018-01-09 02:15

本文关键词：基于时间-小波能量谱样本熵的滚动轴承智能诊断方法　出处：《振动与冲击》2017年09期 　论文类型：期刊论文

【摘要】：为了解决滚动轴承故障模式智能识别与运行状态检测问题,提出了时间-小波能量谱样本熵的计算方法,并将其作为特征参数用于滚动轴承智能诊断的研究。采用Hermitian小波对轴承信号进行连续小波变换,得到蕴含故障信息的时间-小波能量谱序列,再通过计算其样本熵值,量化提取信号中的故障特征信息。轴承不同故障模式下的时间-小波能量谱样本熵区分明显,以此作为特征向量输入支持向量机,实现了对轴承不同故障模式的智能识别。之后计算轴承全寿命周期实验数据的时间-小波能量谱样本熵,按照时间顺序排列,绘制出了轴承运行状态曲线,通过判断曲线走势可有效诊断出轴承早期故障的发生。实验结果表明,时间-小波能量谱样本熵可以有效用于滚动轴承智能诊断的研究。
[Abstract]:In order to solve the fault pattern recognition and state detection of rolling bearing operation, puts forward the calculation method of energy spectrum entropy and wavelet time samples, which are used as feature parameters of the intelligent diagnosis of rolling bearing. The bearing signal using Hermitian wavelet continuous wavelet transform, which get the fault information time wavelet power spectrum sequence then, by calculating the sample entropy, quantitative fault feature extraction of signals. Bearing under different fault time wavelet energy spectrum of sample entropy is obviously different as vectors of support vector machine, realizes the intelligent recognition model of bearing fault. After calculating the data life cycle experiment of bearing time wavelet energy spectrum sample entropy, arranged in chronological order, draw out the bearing running state curve, by judging the trend curve can effectively diagnose the shaft The experimental results show that the time wavelet energy spectrum sample entropy can be effectively used in the research of intelligent diagnosis of rolling bearings.

【作者单位】：石家庄铁道大学机械工程学院;华北电力大学能源动力与机械工程学院;
【基金】：河北省自然科学基金(E2014502052) 中央高校基本科研业务专项资金项目(2014XS83)
【分类号】：TH133.33
【正文快照】： 滚动轴承是旋转机械中支承转动轴的核心部件,对于保持转轴的工作位置和旋转精度具有至关重要的作用,同时它也是机械设备中最易发生故障损坏的部件之一[1]。随着计算机技术的不断发展,采用基于信号处理技术结合人工智能诊断方法对滚动轴承故障进行智能诊断已经越来越广泛。采用

【相似文献】