基于联合分布偏移差异的跨域滚动轴承故障诊断方法

现有的无监督域自适应故障诊断方法大多只基于单一域信号实现,提取的故障信息不够全面。 只注重实现源域和目标域特征的边缘分布对齐,忽略了样本的条件分布差异,限制了诊断精度的提升。 为克服以上问题,提出一种基于联合分布偏移差异(joint distribution offset difference, JDOD)的跨域滚动轴承故障诊断方法。 使用两个结构一致的卷积神经网络(CNN)分别提取信号的时域与频域特征,获取更完整的故障信息。 提出联合分布偏移差异,实现不同域特征的边缘分布对齐和条件分布对齐。 在两个多工况轴承数据集上与多种先进方法展开对比实验,取得了 99%以上的平均诊断精度。 实验结果表明联合分布偏移差异有效提升了跨域故障精度。

Cross-domain fault diagnosis method of rolling bearings basedon joint distribution offset difference

Most of the existing unsupervised domain adaptive fault diagnosis methods are only implemented based on a single domainsignal, and the extracted fault information is not comprehensive enough. Only focus on realizing the edge distribution alignment of sourceand target domain features, ignoring the conditional distribution differences of samples, which limits the improvement of diagnosticaccuracy. To overcome the above problems, a cross-domain fault diagnosis method of rolling bearings based on joint distribution offsetdifferences (JDOD) is proposed. Two structurally consistent CNNs are used to extract the time-domain and frequency-domain features ofthe signal respectively to obtain more complete fault information. Joint distribution offset difference is proposed to realize edge distributionalignment and conditional distribution alignment of different domain features. Comparing experiments with various advanced methods ontwo multi-condition bearing datasets, the average diagnostic accuracy of more than 99% is obtained. The experimental results show thatthe joint distribution offset difference effectively improves the cross-domain fault diagnostic accuracy.