一种基于动态本体演化与异构推理网络的模型训练方法及系统
By employing a model training method based on dynamic ontology evolution and heterogeneous reasoning networks, the problems of static knowledge and unexplainable decisions on edge devices are solved, enabling high-precision fault diagnosis and interpretable reasoning processes, thus adapting to dynamic industrial scenarios.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- XIAMEN MEIYA YIAN INFORMATION TECH CO LTD
- Filing Date
- 2026-04-29
- Publication Date
- 2026-07-17
AI Technical Summary
Existing knowledge augmentation models cannot adapt to dynamic industrial scenarios when deployed on edge devices, resulting in rapid decay of model effectiveness, low resource utilization, and uninterpretable neural network decision-making processes, making it difficult to integrate into industrial processes.
We employ a model training method based on dynamic ontology evolution and heterogeneous reasoning networks to construct a heterogeneous reasoning network that includes a symbolic reasoning module, a neural representation module, and an attention gating module. Through adversarial ontology compression and performance feedback iterative optimization, we achieve the co-evolution of the model and the knowledge base.
Achieving high-precision fault diagnosis on edge devices provides a clear logical chain of evidence, improves the adaptability and resource utilization of the model, and ensures the interpretability of the decision-making process.
Smart Images
Figure CN122114196B_ABST