Additive manufacturing surface topography prediction method and system based on multi-source molten pool feature fusion
By fusing the melt pool feature data collected by a coaxial CMOS camera and a rangefinder near-infrared camera, combined with deep learning technology, the problem of insufficient data from a single sensor is solved, and high-precision surface morphology prediction in the laser additive manufacturing process is achieved, which is suitable for a variety of additive manufacturing processes.
Patent Information
- Application Number
- CN202510749634.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-06
- Publication Date
- 2025-10-10
AI Technical Summary
The data dimension of a single sensor in the existing technology is limited and cannot fully reflect the characteristics and changes of the molten pool. There is a lack of multi-source data fusion methods, and the prediction model is not accurate enough, resulting in insufficient reliability and stability in surface morphology prediction during laser additive manufacturing.
A coaxial CMOS camera and a rangefinder near-infrared camera are used to collect melt pool feature data. Combined with deep learning technology, grid mapping and adaptive multi-source model are used to achieve high-precision prediction of melt pool features and surface morphology.
It realizes real-time and accurate surface morphology prediction in the laser additive manufacturing process, improves prediction accuracy, has wide applicability, reduces costs, and is suitable for a variety of additive manufacturing processes.