Hybrid augmented network with balance domain window for few-shot fault diagnosis under sharp speed variation

摘要

In engineering practice, the operation of mechanical equipment under conditions of sharp speedvariation can result in domain shift in the distribution of samples. Moreover, the presence of datadeficiencies in engineering poses significant challenges for intelligent diagnostic techniques. Toaddress these issues, this paper proposes a network called Hybrid Augmented network withBalance Domain Window (BW-HAN). The BW-HAN network utilizes convolution operations tosegment samples into infinitesimal patches while extracting their underlying features, and thenembeds these patches into attention computational mechanism for invariant domain featureextraction. To handle domain shift, a novel partitioning method is designed based on datarestructuring and data flow. This method aims to suppress domain shift within windows whileestablishing global connectivity between windows. Furthermore, a semi-dense connectionmethod with multi-level residual fusion has been developed based on the principles of featurereuse and parameter sharing to address the issue of overfitting caused by limited samples, therebyenhancing model stability. Comparative studies are conducted with other networks on threevariable-speed cases to demonstrate the superiority of the BW-HAN network. The experimentalresults show that the proposed BW-HAN network performs well in all three cases and effectivelyaddresses the problem of insufficient data under sharp speed variation in mechanical faultdiagnosis.