基于时序知识图谱的复杂工业异常工况因果推理预警方法
By constructing a temporal knowledge graph, quantifying causal strength, and performing causal reasoning, the interpretability and dynamic monitoring of abnormal operating conditions in complex industrial processes are solved, realizing a cross-domain applicable early warning method and improving the safety and efficiency of industrial processes.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG UNIV OF SCI & TECH
- Filing Date
- 2026-05-14
- Publication Date
- 2026-07-17
AI Technical Summary
Existing methods for monitoring and early warning of abnormal industrial conditions lack interpretability, static knowledge representation cannot capture dynamic evolution, causal relationship modeling is insufficient, and knowledge acquisition efficiency is low, making it difficult to effectively monitor and warn of abnormal conditions in complex industrial processes.
A time-series knowledge graph-based approach is adopted. By collecting multi-source heterogeneous time-series data, a static domain knowledge graph with multi-level nodes is constructed, causal strength is quantified, and graph neural networks are used for causal reasoning to generate interpretable early warning information.
It enables interpretable early warning for complex industrial processes, can monitor and locate root causes online, and supports complete decision support from post-event alarms to pre-event warnings, with cross-domain universality and engineering applicability.
Smart Images
Figure CN122200957B_ABST