Urban rail fault root cause automatic analysis method and system based on industry knowledge graph and large model

By constructing an automatic root cause analysis system for urban rail transit faults based on knowledge graphs and large language models, the problems of knowledge scarcity and manual dependence on unstructured data in operation and maintenance have been solved, achieving efficient and accurate root cause analysis and adaptive decision support.

CN122365192APending Publication Date: 2026-07-10INTELLIGENT & TECHNOLOGY BRANCH OF CHINA RAILWAY FIRST GROUP CO LTD
0 Cites 0 Cited by

Patent Information

Application Number
CN202610452203.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-04-08
Publication Date
2026-07-10

AI Technical Summary

Technical Problem

In urban rail transit operation and maintenance, unstructured data results in abundant data but scarce knowledge. Diagnosis relies on human experience and lacks standardized decision support. Existing monitoring systems cannot provide complete root cause analysis of faults and maintenance solutions.

Method used

A multi-source heterogeneous data processing system is constructed using an industry knowledge graph and large language model approach to achieve automatic root cause analysis of faults. This includes data acquisition, knowledge graph construction, model fine-tuning, inference engine construction, and human-machine feedback mechanism to generate fault reports and maintenance plans.

Benefits of technology

It enables the transformation from unstructured data to structured knowledge, improves the accuracy and efficiency of root cause analysis, reduces reliance on manual intervention, and provides complete fault decision support and adaptive optimization capabilities.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses an urban rail fault root cause automatic analysis method and system based on an industry knowledge graph and a large model. According to the method, knowledge is automatically extracted from unstructured historical maintenance records and technical manuals, and a special knowledge graph for urban rail operation and maintenance is constructed; and automatically generating a natural language diagnosis report containing a fault root cause, a risk level, a maintenance scheme and a spare part list by utilizing the fine-tuned vertical field large language model and combining real-time monitoring data and a map retrieval result. According to the method, the problems that expert experience digitization is difficult, and fault disposal depends on manual consulting in urban rail operation and maintenance are effectively solved, the intelligent level, accuracy and efficiency of fault diagnosis are remarkably improved, and decision closed-loop support of operation and maintenance is achieved.
Need to check novelty before this filing date? Find Prior Art