Equipment fault prediction system based on multi-modal deep learning

CN119538185AActive Publication Date: 2025-02-28HUAZHONG UNIV OF SCI & TECH
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411571509.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-06
Publication Date
2025-02-28
Estimated Expiration
2044-11-06

AI Technical Summary

Technical Problem

In the prior art, the model-based equipment fault prediction accuracy is not high, and it is difficult to effectively predict the health status of the equipment.

Method used

Using a device fault prediction system based on multimodal deep learning, the image features are extracted through the image processing model and fused with external timing vectors, input into the TranAD model for prediction, and the failure risk is judged through error reports.

Benefits of technology

Improve the accuracy of equipment failure prediction. Through multimodal deep learning, the health status of the device can be more accurately judged and more reliable fault prediction can be provided through error calculation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119538185A_ABST
    Figure CN119538185A_ABST
Patent Text Reader

Abstract

The invention belongs to the related technical field of equipment fault prediction, and discloses an equipment fault prediction system based on multi-modal deep learning, which is characterized by comprising a picture processing model, a multi-modal deep learning model, a multi-modal deep learning model, a multi-modal deep learning model, a multi-modal deep learning model and a multi-modal deep learning model, the encoder extracts a picture feature vector of an external input picture and inputs the picture feature vector into the generator to reconstruct the picture, the first decoder performs feature extraction on the external input picture to obtain a first decoding vector, and the second decoder performs feature extraction on the reconstructed picture to obtain a second decoding vector; the fusion module is used for carrying out feature fusion on the external time sequence vector and the picture feature vector, constructing a feature matrix and inputting the feature matrix into a TransAD model; the TransAD model is used for predicting feature data in the next time window based on the feature matrix in the current time window; and the report output module is used for calculating errors and outputting error reports for fault prediction, and the larger the errors are, the higher the probability that the equipment has fault risks is. Through the method, the fault prediction precision can be improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Mechanical arm autonomous operation strategy learning method based on vision-touch fusion

    CN114660934A

  • Zero-sample composite fault diagnosis method based on multi-modal contrast embedding

    CN117909854A

  • Apparatus and method for processing multi-type sensor signal on the basis of multi-modal deep learning

    US20190293462A1

  • Multi-modal deep learning power generation device anomaly integrated identification method and device

    WO2023087525A1