Rotary machinery fault diagnosis method based on adaptive wavelet feature fusion network

CN119312280APending Publication Date: 2025-01-14SHANGHAI ELECTRICGROUP CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411380590.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-30
Publication Date
2025-01-14

AI Technical Summary

Technical Problem

In the fault diagnosis of rotating machinery, the strong noise signal processing performance is insufficient, resulting in increased difficulty in identifying fault types and requires profound theoretical knowledge and practical experience.

Method used

The fault diagnosis method based on the adaptive wavelet feature fusion network is adopted, and the convolution layer replaces the update operator and prediction operator in the wavelet transform, high-frequency and low-frequency features are extracted and fused, combining cross-fusion technology and channel attention mechanism , build an adaptive improvement solution module to enhance the feature extraction capability and robustness of the fault diagnosis model.

Benefits of technology

It improves the feature extraction accuracy and noise robustness of the fault diagnosis model, enhances the understanding of complex signals and noise resistance, provides new research ideas and technical methods, and significantly improves the accuracy of rotary machinery fault diagnosis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119312280A_ABST
    Figure CN119312280A_ABST
Patent Text Reader

Abstract

The invention discloses a fault diagnosis method for a rotating machine based on a self-adaptive wavelet feature fusion network. The fault diagnosis method comprises the following steps: acquiring a vibration signal of the rotating machine; inputting the vibration signal of the rotating machine into a preset adaptive wavelet feature fusion network diagnosis model to obtain a fault diagnosis result; the adaptive wavelet feature fusion network diagnosis model uses a convolutional layer to replace an update operator and a prediction operator in wavelet transform so as to extract high-frequency features and low-frequency features, and the high-frequency features and the low-frequency features are fused to obtain fused features; and carrying out fault classification based on the fused features. According to the fault diagnosis method, a traditional lifting scheme is combined with the neural network, so that the neural network can more accurately capture key features of strong noise signals; through a joint cross feature fusion mechanism, features of different levels are fused, and the understanding of a fault diagnosis model for complex signals and the robustness of noise are enhanced.
Need to check novelty before this filing date? Find Prior Art