基于双维度特征的输氢管网流量异常检测方法及系统
By employing adaptive slicing with dual-dimensional features and multi-view reconstruction networks in hydrogen pipeline networks, the problem of insufficient utilization of inter-sensor relationships was solved, enabling accurate anomaly detection of hydrogen pipeline network flow and improving safety management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG COMP SCI CENTNAT SUPERCOMP CENT IN JINAN
- Filing Date
- 2024-08-27
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies fail to fully utilize the interrelationships between sensors in hydrogen pipeline network flow monitoring, resulting in a high false alarm rate when a single sensor fails, increasing operation and maintenance costs, and lacking flexibility and accuracy.
A method for detecting flow anomalies in hydrogen pipeline networks based on dual-dimensional features is adopted. By adaptive slicing in the time and channel dimensions, a time feature extractor, a spatial feature extractor, and a trunk feature extractor are used to capture the feature correlations within and between slices. Finally, a reconstruction network with multiple hidden layers of different feature scales is used to reconstruct and score anomalies under multiple views.
It enables accurate identification of abnormal flow data in industrial hydrogen transmission pipelines, improves industrial safety management, and ensures the stable operation of hydrogen transmission pipelines.
Smart Images
Figure CN119150186B_ABST