Research On Defect Detection And Compression Performance Of 3D Printing Lattice Structures

Specific metal lattice structures can be printed quickly and efficiently by 3D printing technology,which can meet the urgent demand for lightweight metal lattice structures in the aerospace field.The defects of lattice structure such as warpage,crack and geometric fracture will greatly reduce the structure-functional characteristics of lattice structure and affect the safe use of lattice structure.Therefore,it is of great significance to develop an internal defect detection method for 3D printed metal lattice structures and research on their structural properties.In this paper,the defect detection and static mechanical finite element analysis of titanium alloy lattice structures prepared by Selective Laser Melting(SLM)technology have been carried out.The specific research contents are as follows:(1)Considering the complexity of metal lattice structure and the randomness,diversity of features and invisibility of internal defects,the gray scale characteristics and morphological characteristics of defects in industrial CT images with lattice structure were analyzed,and the difference law between defect area and non-defect area was given.(2)According to the distribution characteristics of gray difference in the area with or without defects in CT images,a method of automatic defect detection by sliding window is proposed.The window size was determined by the gray distribution law of the image,the sliding window was used to traversal the image,and the influence of gray inhomogeneity was eliminated by the difference principle.The distribution law of the upper and lower edges of the defect area in the difference image was analyzed,and the defect identification rules were established.(3)Aiming at the problems of small defect scale,high background recognition and weak features in metallic lattice structure,a defect detection method based on super resolution reconstruction is proposed.In order to solve the problem of low recognition rate of small-scale target by Faster R-CNN model,the original backbone feature network is deepened,and the image super-resolution reconstruction module is developed to enhance local detail features of the input image,so as to realize fast and effective location recognition of two kinds of different micro-defects(4)On the basis of defect detection,focusing on the influence of internal defects on lattice structure performance,a static mechanical model was established based on numerical simulation analysis method.Considering the influence of the position and number of defects,the failure process of titanium alloy lattice structure with different position and number of defects is simulated and analyzed,and the influence of defects on the failure process of lattice structure and the variation law of yield load are revealed,which provides a technical basis for the engineering application of lattice structure.