一种数控机床的AI驱动型机器视觉刀具状态异常预警方法
By using AI-driven machine vision methods, cross-material adaptation and bidirectional correction are performed on the digital twin simulation engine and chip reverse engineering network of CNC machine tools. This solves the problem of accuracy in tool status monitoring under new material tool combinations for CNC machine tools and achieves reliable anomaly early warning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUNAN VOCATIONAL COLLEGE OF SCI & TECH
- Filing Date
- 2026-04-27
- Publication Date
- 2026-07-17
AI Technical Summary
In the existing technology, when CNC machine tools introduce new alloy materials or new coated tool combinations, the tool condition monitoring methods based on digital twin simulation engines and machine vision operate independently and cannot detect migration errors on their own. This results in missed or false alarms during the cold start phase of the new process, and cannot provide reliable early warning of abnormal tool conditions.
By introducing AI-driven machine vision methods, the attribute description vector of the new material tool combination is obtained. The parameter transfer network and material property conditional feature transformation are used to adapt the digital twin simulation engine and chip inverse network across materials. The bidirectional collaborative correction is performed through cross-bias calculation and online Bayesian parameter update to generate a consistent estimate of the crescent depression depth.
It enables accurate monitoring and timely early warning of tool status under the new material tool combination, reduces missed and false alarms, and improves the reliability of the early warning system.
Smart Images
Figure CN122077451B_ABST