基于敏感模态流形的变转速故障诊断方法

Variable Speed Fault Diagnosis Method Based on Sensitive Modal Manifold

针对旋转机械变转速和复杂工况多分类问题,提出了一种基于构建改进敏感模态矩阵(ISMM)、等度量映射(ISOMAP)和convolution-vision transformer(CVT)网络结构的故障诊断方法。将变转速信号重叠采样之后构造高维ISMM,通过ISOMAP流形学习将ISMM映射到流形空间中,提取变转速信号的故障瞬态特征,实验证明能够很好地解决了常规方法无法对变转速数据进行有效特征提取的问题。结合自注意力机制和CNN的优点,使用CVT网络结构进行特征提取、故障识别分类。通过实验室HFXZ-Ⅰ行星齿轮箱变转速数据集对提出的故障诊断模型进行实验验证。实验结果表明,提出的故障诊断模型具有良好的识别准确率及鲁棒性。

Rotating machinery(RM)is one of the most common mechanical devices with a wide and critical role in engineering applications.A fault diagnosis method based on the construction of improved sensitive modal matrix(ISMM),ISO-metric mapping(ISOMAP)and convolution-vision transformer(CVT)network structure is proposed for the problem of variable speed and complex working conditions of rotating machines with multiple classifications.After overlapping sampling of the variable speed signal,a high-dimensional ISMM is constructed,and the ISMM is mapped into the streaming space through ISOMAP stream learning to extract the fault transient features of the variable speed signal.Combining the advantages of the self-attention mechanism and CNN,the CVT network structure is used for feature extraction,fault identification,and classification.The laboratory HFXZ-Ⅰplanetary gearbox variable speed data set experimentally validates the proposed fault diagnosis model.The experimental results show that the proposed fault diagnosis model has good recognition accuracy and robustness.