A self-service diagnosis method for power system faults based on conversational reasoning and generation
By introducing dialogue inference and knowledge generation capabilities of large language models in power system fault diagnosis, the lack of dynamic inference, strong data dependence and insufficient interpretation in the existing technology is solved, and more efficient, flexible and intelligent fault diagnosis is achieved.
Patent Information
- Application Number
- CN202411541328.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-31
- Publication Date
- 2025-04-01
AI Technical Summary
The existing power system fault diagnosis methods lack dynamic inference capabilities, have strong data dependence, insufficient interpretation and interactivity of diagnostic results, and cannot effectively process multimodal data.
Introduce dialogue-based reasoning and knowledge generation capabilities of large language models, dynamically adjust the inference path through multiple rounds of interactions, introduce external knowledge for fault analysis, and realize comprehensive analysis and comprehensive diagnosis of multimodal data.
It improves the intelligence, flexibility and interpretation of power system fault diagnosis, can provide accurate diagnostic results in complex and changeable environments, and enhances the adaptability and maintainability of the system.