小波包-局部线性嵌入算法在滚动轴承故障程度识别中的应用

发布时间：2018-02-24 09:38

本文关键词： 小波包分解局部线性嵌入支持向量机滚动轴承性能退化　出处：《仪器仪表学报》2014年03期 　论文类型：期刊论文

【摘要】：对滚动轴承不同性能退化程度进行识别是性能退化评估的前提。为了提高性能退化程度的识别准确率,提出一种基于小波包与局部线性嵌入算法相结合的滚动轴承故障性能退化程度识别方法。首先计算单个样本振动信号的时域、频域指标,再对振动信号进行小波包分解并重构,提取节点信号的时域、频域指标和奇异值构造特征向量,并将多个样本的特征向量排列组合成特征矩阵,然后利用局部线性嵌入算法对特征矩阵当中的特征维数进行约简。最后利用支持向量机对滚动轴承不同性能退化程度进行识别,验证提出方法的可行性和有效性。实验表明,该方法可以准确地识别出滚动轴承不同性能退化程度。
[Abstract]:The recognition of different performance degradation degree of rolling bearing is the premise of performance degradation assessment. In order to improve the recognition accuracy of performance degradation degree, A method for identifying the degree of fault degradation of rolling bearing based on wavelet packet and local linear embedding algorithm is proposed. Firstly, the time-domain and frequency-domain indexes of single sample vibration signal are calculated, then the vibration signal is decomposed and reconstructed by wavelet packet. The time domain, frequency domain index and singular value of the node signal are extracted to construct the eigenvector, and the Eigenvectors of several samples are arranged and combined into a feature matrix. Then the local linear embedding algorithm is used to reduce the feature dimension in the feature matrix. Finally, support vector machine is used to identify the different performance degradation degree of the rolling bearing, which verifies the feasibility and effectiveness of the proposed method. This method can accurately identify the different performance degradation degree of rolling bearing.
【作者单位】：哈尔滨理工大学电气与电子工程学院;哈尔滨工业大学电子与信息工程学院;
【基金】：国家自然科学基金(51305109) 高等学校博士学科点专项科研资助(20122303120010) 哈尔滨市科技创新人才专项基金(留学回国人员)(2013RFLXJ019)资助项目
【分类号】：TH165.3;TN911.7

【参考文献】