Dynamic penalty adaptive matrix machine for the intelligent detection of unbalanced faults in roller bearing

作者：

Highlights：

• A novel dynamic penalty adaptive matrix machine (DPAMM) is proposed for classification of unbalanced samples.

• The adaptive low-rank approximation minimization framework is used to adaptively select the larger singular values related to the strong correlation information within the matrix samples.

• The dynamic penalty factor is introduced to establish a classification model, which can weaken the influence of unbalanced samples on model training and improve the classification performance.

摘要

•A novel dynamic penalty adaptive matrix machine (DPAMM) is proposed for classification of unbalanced samples.•The adaptive low-rank approximation minimization framework is used to adaptively select the larger singular values related to the strong correlation information within the matrix samples.•The dynamic penalty factor is introduced to establish a classification model, which can weaken the influence of unbalanced samples on model training and improve the classification performance.

论文关键词：Dynamic penalty adaptive matrix machine,Adaptive low-rank operator,Dynamic penalty factor,Unbalanced samples,Fault diagnosis

论文评审过程：Received 9 February 2022, Revised 7 April 2022, Accepted 7 April 2022, Available online 18 April 2022, Version of Record 28 April 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2022.108779