Search

Aiming at the problem of the poor robustness of existing methods to deal with diverse industrial weld image data, we collected a series of asymmetric laser weld images in the largest laser equipment workshop in Asia, and studied these data based on an industrial image processing algorithm and deep learning algorithm. The median filter was used to remove the noises in weld images. The image enhancement technique was adopted to increase the image contrast in different areas. The deep convolutional neural network (CNN) was employed for feature extraction; the activation function and the adaptive pooling approach were improved. Transfer Learning (TL) was introduced for defect detection and image classification on the dataset. Finally, a deep learning-based model was constructed for weld defect detection and image recognition. Specific instance datasets verified the model's performance. The results demonstrate that this model can accurately identify weld defects and eliminate the complexity of manually extracting features, reaching a recognition accuracy of 98.75%. Hence, the reliability and automation of detection and recognition are improved significantly. The research results can provide a theoretical and practical reference for the defect detection of sheet metal laser welding and the development of the industrial laser manufacturing industry. [ABSTRACT FROM AUTHOR]

The emergence of computer-aided design (CAD) has propelled the evolution of the sheet metal engineering field. Sheet metal design software tools include parameters associated to the part's forming process during the pattern drawing calculation. Current methods avoid the calculation of a first pattern drawing of the flattened part's neutral surface, independent of the forming process, leading to several methodological limitations. The study evaluates the reliability of the Computer Extended Descriptive Geometry (CeDG) approach to surpass those limitations. Three study cases that cover a significative range of sheet metal systems are defined and the associated solid models and patterns' drawings are computed through Geogebra-based CeDG and two selected CAD tools (Solid Edge 2020, LogiTRACE v14), with the aim of comparing their reliability and accuracy. Our results pointed to several methodological lacks in LogiTRACE and Solid Edge that prevented to solve properly several study cases. In opposition, the novel CeDG approach for the computer parametric modeling of 3D geometric systems overcame those limitations so that all models could be built and flattened with accuracy and without methodological limitations. As additional conclusion, the success of CeDG suggests the necessity to recover the relevance of descriptive geometry as a key core in graphic engineering. [ABSTRACT FROM AUTHOR]

This article introduces an innovative application of Visual Basic for Applications (VBA) (Visual Basic macro language) developed in CATIA (Computer-Aided Three-Dimensional Interactive Application), computer-aided manufacturing and engineering design software. Said application enables the three-dimensional (3D) representation and development to be obtained on a plane of the surfaces of cylindrical and conical elbows utilised to join two conduits, since CATIA lacks tools for the design of elements of metal fabrication. The application permits the geometric and dimensional parameters necessary for the creation and development of the analysed elbows to be introduced via a keyboard. Once these values have been entered and it has been verified that the design of the elbow is possible, then the application is programmed within the CATIA "Wireframe and Surface Design" module to obtain the 3D representation of said elbows, and the development of their surface is projected in the CATIA "Drafting" module. Consequently, the incorporation of this tool into CATIA software could increase technical-design productivity by eliminating certain intermediate operations, such as symmetry, and through improving the geometry training of less skilled users. [ABSTRACT FROM AUTHOR]