A substation switch port flow anomaly diagnosis method and device

By building a dynamic baseline model and differentiated threshold strategy in substation switch monitoring, the problem that static threshold models cannot adapt to dynamic changes in business is solved, enabling accurate anomaly diagnosis and risk assessment, reducing deployment complexity, and improving the availability and fault handling efficiency of the monitoring system.

CN122179402APending Publication Date: 2026-06-09GUIZHOU ANRONG TECH DEV CO LTD +2
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
GUIZHOU ANRONG TECH DEV CO LTD
Filing Date
2026-05-09
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

Existing substation switch monitoring technologies suffer from problems such as static threshold models being unable to adapt to dynamic changes in business operations, lack of fine-grained anomaly diagnosis capabilities, difficulty in quickly associating business impacts, and high deployment complexity.

Method used

High-frequency data acquisition based on substation network topology information and SCD configuration files is adopted. A dynamic baseline model is constructed by combining matrix tree theorem and sliding time window algorithm. Risk cloud map and fault isolation suggestions are generated through differentiated threshold strategy and abnormal event correlation analysis.

Benefits of technology

It enables accurate anomaly diagnosis of substation switch port traffic, improves detection accuracy and efficiency, reduces deployment complexity, adapts to dynamic business changes, and provides accurate risk assessment and fault isolation suggestions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122179402A_ABST
    Figure CN122179402A_ABST
Patent Text Reader

Abstract

The application discloses a transformer substation switch port flow abnormality diagnosis method and device, high-frequency data of the switch is collected through an SNMP probe, a SCD configuration file is analyzed to establish a mapping relationship between communication nodes and switch ports, and a physical network topology graph is constructed; the number of spanning trees of each service node is calculated as an index of network connectivity contribution degree by using a matrix tree theorem, and the importance of services borne by each port is rated; a sliding time window algorithm is used to extract time sequence characteristics, and a global minimum cut approximation algorithm is combined to establish a dynamic baseline model of each port; a differentiated threshold strategy is used for abnormality detection on ports of different service levels; an original dual covering small cut method is used for abnormal event correlation analysis, and a three-layer correlation model of a port-link-protection interval is established. The problems that a traditional static threshold model cannot adapt to dynamic changes of services are solved, and accurate mapping from network abnormality to service influence is realized.
Need to check novelty before this filing date? Find Prior Art