Robustness Imbalanced Convex Hull-based Classification for Bevel Gearbox Fault Diagnosis

Xin Li, Yang Yu, Jian Cheng, Junsheng ChengJournal of Mechanical Engineering 2023 · DOI

Abstract

摘要: 实际工况下,拾取的锥齿轮箱振动信号中不可避免会掺杂噪声及异常点。同时,考虑到故障样本获取困难,提出一种鲁棒不平衡凸包分类(Robustness imbalanced convex hull-based classification,RICHC)模型用于锥齿轮箱故障智能诊断。RICHC根据不同样本在类别分布估计中的作用,确定各样本的置信函数,降低异常值和含噪样本的权重,使凸包模型边界更加紧致,以提高模型的鲁棒性。同时,构建自适应模型缩放策略,使RICHC根据多数类和少数类间的动态不平衡因子,调整不同类别间凸包的缩放比例,得到更加准确的分类超平面,从而提升模型的不平衡数据处理能力。采用锥齿轮箱故障数据验证所提方法的有效性及适用性,试验结果表明:相较于其他算法,所提方法对噪声和异常值具有更强的抗干扰能力,且该方法具有更优异的类不平衡分类性能。
BibTeX
@article{li2023robustness,
  author = {Xin Li and Yang Yu and Jian Cheng and Junsheng Cheng},
  title = {{Robustness Imbalanced Convex Hull-based Classification for Bevel Gearbox Fault Diagnosis}},
  journal = {Journal of Mechanical Engineering},
  year = {2023},
  doi = {10.3901/jme.2023.08.032},
}

Loading related content...

Figure