Mechanical fault diagnosis method based on multi-source information self-integration neural network

By adopting a method based on multi-source information self-integrated neural network in mechanical fault diagnosis, building a hybrid expert network and combining the multi-layer maximum mean difference method, the limitations of the multi-source domain adaptation method in fault diagnosis are solved, and more efficient fault diagnosis accuracy and robustness are achieved.

CN120145139AActive Publication Date: 2025-06-13苏州大珩数智科技有限公司
4 Cites 2 Cited by

Patent Information

Application Number
CN202510188907.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-02-20
Publication Date
2025-06-13
Estimated Expiration
2045-02-20

AI Technical Summary

Technical Problem

The existing multi-source domain adaptation methods have limitations in fault diagnosis, including the limitations of multi-source domain selection and fusion. Output weighting depends on a wide range of domain-level operations, the output combination lacks consistency guarantee, knowledge distillation is difficult to extract accurate knowledge, multi-adversarial learning cannot completely resolve conflicts, feature selection cannot capture sufficient informational features, and weighted learning is difficult to adapt to different degrees of domain differences.

Method used

The mechanical fault diagnosis method based on multi-source information self-integrated neural network is adopted. By building a hybrid expert network, the shared feature extractor extracts the domain-invariant features across domains. The domain-specific feature extractor further processes domain-specific features, combines the multi-layer maximum mean difference method to achieve domain alignment, and uses a comprehensive loss function to train the model to balance the trade-offs of different loss terms.

Benefits of technology

Improve the accuracy and efficiency of fault diagnosis, and the hybrid expert model of learning distance measurement flexibly weighted the characteristics of multiple source domains, achieving the accuracy and robustness of cross-domain fault diagnosis.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a mechanical fault diagnosis method based on a multi-source information self-integration neural network, and the method comprises the steps: sequentially selecting samples from each source domain, and calculating a point pair set distance set of the samples relative to a plurality of source domains, so as to construct a positive sample pair and a negative sample pair; establishing a first loss function for reducing the point pair set distance measure of the positive sample pair and increasing the point pair set distance measure of the negative sample pair, and training to obtain the point pair set distance measure; establishing a second loss function including multi-task learning loss and joint training loss, and training to obtain a hybrid expert network; establishing a third loss function for aligning the low-level features and the high-level features of the cross-domain through the multi-layer maximum mean value difference, and aligning the input and output features of the hybrid expert network; and training the aligned hybrid expert network to obtain a learnable distance measurement hybrid expert model for fault diagnosis. According to the method, specific feature output of multiple source domains can be integrated, gating output matching and independent task learning are balanced, and complex nonlinear domain drift can be processed.
Need to check novelty before this filing date? Find Prior Art