Crack real-time monitoring method, device and equipment in additive manufacturing process and medium

By generating a thermal map from the molten pool radiation signal online and combining it with CT scan image data, a layer-by-layer, point-by-point dataset is established. A machine learning model is used to realize real-time crack monitoring in the laser additive manufacturing process, which solves the problem of unstable crack monitoring in the existing technology and improves the quality and efficiency of the formed parts.

CN122109103APending Publication Date: 2026-05-29UNIV OF SCI & TECH BEIJING
6 Cites 0 Cited by

Patent Information

Application Number
CN202610326103.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-17
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing technologies make it difficult to achieve real-time monitoring of cracks during laser additive manufacturing, especially for large-sized formed parts where fine three-dimensional scanning is difficult. Furthermore, existing online monitoring solutions lack processing rules for generating radiation signal thermograms with uniform resolution from raw monitoring data, and the data correspondence is unstable, resulting in poor model transferability.

Method used

By collecting molten pool radiation signals online, a radiation signal heat map is generated. Combined with CT scan image data, a radiation signal crack feature dataset corresponding to each layer and point is established. Machine learning or deep learning models are used to determine cracks, enabling real-time monitoring and location of cracks.

Benefits of technology

It enables real-time monitoring and precise positioning of cracks during laser additive manufacturing, improving the quality and manufacturing efficiency of formed parts, and supporting online rapid control and digital/intelligent additive manufacturing.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to a real-time crack monitoring method, device, equipment and medium in the additive manufacturing process, and the method comprises the following steps: S1, collecting a laser radiation signal for a formed part in real time in the additive manufacturing process, converting the laser radiation signal, and processing the converted electric radiation characteristic signal to obtain a radiation signal thermodynamic diagram of the formed part; s2, CT scanning is carried out on the formed part at the same time to obtain image data, and a crack characteristic distribution diagram is obtained after processing; s3, the radiation signal thermodynamic diagram and the crack characteristic distribution diagram correspond to the number of layers of the formed part, and a layer-by-layer and point-by-point corresponding radiation signal crack characteristic data set is obtained; s4, training a machine learning or deep learning model by adopting the data set to obtain a crack judgment model; s5, the crack judgment model is used for on-line monitoring for crack recognition. According to the method disclosed by the invention, a layer-by-layer and point-by-point corresponding radiation signal crack database in a three-dimensional space can be formed, and the data consistency and mobility are improved.
Need to check novelty before this filing date? Find Prior Art