An abnormal working condition intelligent prediction diagnosis method in logistics line butt joint process

By combining multi-source sensors and intelligent algorithms, the system generates criteria for judging abnormal behavior and conducts confidence level assessments, solving the problem of difficulty in identifying anomalies during logistics line docking. This enables accurate anomaly diagnosis and control, improving the stability and security of the logistics docking process.

CN122116271APending Publication Date: 2026-05-29GUANGDONG JIUYING PRECISION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
GUANGDONG JIUYING PRECISION TECH CO LTD
Filing Date
2026-02-11
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Anomaly identification during logistics line docking is difficult, and traditional methods are insufficient for effective and accurate diagnosis and control.

Method used

Image annotation data is acquired through multi-source sensors, and occlusion prediction and illumination compensation are used to dynamically repair transmitted images in real time. An attention-guided deformable convolutional network is designed, and spatiotemporal constraint matching is performed in combination with a docking rule knowledge graph to generate criteria for judging abnormal behavior. An abnormal probability is calculated through a confidence evaluation model, and a confidence-driven bidirectional feedback mechanism and a self-optimizing closed-loop verification system are deployed to achieve continuous enhancement of algorithm iteration and judgment logic.

Benefits of technology

It enables autonomous identification and classification of abnormal behaviors during logistics line docking, ensuring the accuracy of identification results, timely detection of anomalies, avoiding omissions, and improving the stability and security of the logistics docking process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122116271A_ABST
    Figure CN122116271A_ABST
Patent Text Reader

Abstract

The application discloses an abnormal working condition intelligent prediction diagnosis method in a logistics line butt joint process, relates to the technical field of logistics, and comprises the following steps: generating a transmission enhanced image; designing an attention guided deformable convolution network; extracting key frames of a butt joint main body motion track through a real-time heat distribution map; combining a butt joint rule knowledge graph to perform space-time constraint matching, generating abnormal behavior judgment basis with an abnormal label; calculating an abnormal probability through a confidence evaluation model and generating a visual judgment result; triggering manual review when the confidence is low, and synchronously feeding back the result to a compensation model and a reasoning engine; dynamically updating convolution network parameters and constraint matching weight coefficients, and realizing algorithm iteration and continuous enhancement of judgment logic. Through the generation of abnormal behavior judgment basis with an abnormal label and the calculation of an abnormal probability and the generation of a visual judgment result, identification and classification can be performed, the confidence of the identified result is evaluated, and logistics butt joint abnormalities can be found in a timely manner.
Need to check novelty before this filing date? Find Prior Art