一种微组装缺陷可解释性分析系统及方法
By combining data acquisition and feature quantization with a physical rule knowledge base and Bayesian networks, the interpretability problem of micro-assembly defect analysis was solved, enabling accurate analysis of zero-sample and composite defects, and improving the robustness and efficiency of the production system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- PHASYM TECH CO LTD
- Filing Date
- 2025-12-04
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies rely on experience for defect cause analysis in micro-assembly processes, lack interpretability, struggle to handle zero-sample or complex defects, and existing knowledge graphs cannot effectively decouple the interactions between multiple process stages.
The system uses a data acquisition subsystem to acquire detection data, a feature extraction subsystem to quantify defect features, and a physical rule knowledge base and a Bayesian network to perform probabilistic reasoning to determine the root cause of the process and generate an interpretable diagnostic report.
It enables effective analysis of zero-sample defects, accurately decouples the causes of complex defects, reduces trial-and-error costs and optimization cycles, and improves the robustness and adaptability of the production system.
Smart Images
Figure CN121684080B_ABST