TY - JOUR A2 - Pan, Haiyang AU - Sun, Guodong AU - Hu, Ye AU - Wu, Bo AU - Zhou, Hongyu PY - 2021 DA - 2021/05/19 TI - Rolling Bearing Fault Diagnosis Method Based on Multisynchrosqueezing S Transform and Faster Dictionary Learning SP - 8456991 VL - 2021 AB - Addressing the problem that it is difficult to extract the features of vibration signal and diagnose the fault of rolling bearing, we propose a novel diagnosis method combining multisynchrosqueezing S transform and faster dictionary learning (MSSST-FDL).优先使用MSSST将振荡信号转换为高分辨率时频图像本地二元模式运算符取出时频图像低维纹理特征,提高故障识别速度最后,非负矩阵分解法(NMF)只有一个超参数和非负线性方程分别用于解决词典学习和特征编码问题特征编码输入分类器培训识别实验显示,我们方法在Case西部保留大学和机械故障预防技术学会滚动带数据集方面表现良好此外,拟方法应用到扩音器纯调检测数据集中,并实现扩音器异常诊断诊断结果验证我们方法能满足实用工程需求SN-1070-9622UR-https://doi.org/101155/2021/856991DO-10.1155/2021/856991JF-震荡和振荡PB-HindawiKW-ER-