Method for identifying defects of metal additive manufacturing workpiece and related equipment

Abstract

The embodiment of the invention discloses a method for identifying defects of a metal additive manufacturing workpiece and related equipment. The method comprises the following steps: acquiring laser-induced plasma characteristic spectrums of at least two metal elements on the metal additive manufacturing workpiece by utilizing a laser induction technology; acquiring a spectral data matrix of the metal additive manufacturing workpiece based on the laser-induced plasma characteristic spectrum; performing dimension reduction processing on the spectral data matrix in a principal component analysis mode to obtain a target number of principal components; and substituting the target number of principal components as input variables into a random forest model based on a decision tree classifier to obtain a defect identification result of the metal additive manufacturing part. Through the technical scheme of the embodiment of the invention, real-time, nondestructive and multi-element online analysis of the defects of the metal additive manufacturing part can be realized, and the efficiency and the accuracy of the defects of the metal additive manufacturing part are improved.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of detection of metal additive manufacturing parts, and in particular, to a method for identifying defects in metal additive manufacturing parts and related equipment. Background technique [0002] Additive Manufacturing (AM for short, also known as 3D printing technology) has developed rapidly in the field of metal material applications and has made great achievements. AM technology is different from the traditional subtractive manufacturing process, which needs to cut raw materials and assemble parts, etc., but integrates materials processing, sample molding and computer-aided technologies from scratch and "bottom-up" through material accumulation. manufacturing method. At the same time, AM technology has the flexibility of internal complex structures that cannot be manufactured by traditional machinery, and the possibility of growing parts with required chemical composition gradient...

AI Technical Summary

Problems solved by technology

Among them, the offline detection methods include CT, osmosis, eddy current and ultrasonic detection, etc., which require complex sample preparation and a large amount of analysis time, and cannot take advantage of the short manufacturing cycle of AM technology.

On-line detection methods include optical temperature measurement, high-speed photography, and high-speed X-photography imaging. On-line inspection methods also have certain disadvantages in defect detection. For example, high-speed photography can only detect the state of the molten pool macroscopically, but cannot detect a small area. Defects

Images