Gas flow status analysis in CMT+P additive manufacturing based on texture features of molten pool images

Abstract

A new method based on gray level co-occurrence matrix (GLCM) was proposed to extract the texture features of molten pool images and used to monitor the gas flow status in the cold metal transfer plus pulse (CMT + P) based additive manufacturing. The current and intensity signals of CMT + P in the short circuit stage were collected and analyzed to get clearer molten pool images. According to the characteristics of GLCM, the best parameter group was identified through comparative analysis, and four texture features of angular second moment, entropy, contrast and correlation were extracted. On this basis, the gas flow status was divided into three categories of LOW, MEDIUM and HIGH, which were identified successfully using a prediction model based on support vector machine and cross validation.