AUV Fault Diagnosis Model Based on Multi-source Signal Fusion and Deep Residual Shrinkage Network
XU Wei
ZHU Zhiyu
WEI Lai
ZHU Dewen
ZANG Xu
Abstract：Single sensor signals are susceptible to interference from external noise and may not be sufficient to accurately char-acterise the health of the AUV thruster,leading to uncertainty in fault diagnosis.To this end,this paper proposes an AUV thruster fault diagnosis model based on multi-source signal fusion and deep residual shrinkage network.The model has three independent channels,and the angular velocity signals collected by gyroscopes in different directions are used as the input vectors of each chan-nel,the residual shrinkage network is applied to adaptively remove the noise interference of each sensor signal and extract useful fault features.Finally,a fully connected layer is introduced to fuse the features of each channel signal for diagnosis and output the diagnosis results.The experimental results show that the multi-source signal fusion diagnosis model proposed in this paper achieves 100%accuracy,effectively eliminates the uncertainty in the diagnosis process,and has good noise immunity in the case of small samples,and has the feasibility of practical application.
Keywords：AUVfault diagnosismulti-source signal fusionsmall samplesdeep residual shrinkage networks
Publication Date：2025-09-20
Online Publishing Date：2025-12-23(First online date of this platform, not the publication date of the document)
Pages：6( 2455-2460 )