Automatic Extraction and Welding Feature Recognition from STEP Data

Feature recognition has been considered as an important bridge between CAD and CAPP in process planning for welding. These recognition data can be used as input data not only for generating welding tasks in process planning but also for programming work of welding industrial robot without using teach pendant. This paper introduces an effective method to extract and recognize welding features from STEP file which can be provided by any 3D CAD system. The computer program quickly extracts data from various strings in the STEP file, saves them in the database and recognizes features for welding by the developed algorithm. The welding paths with detailed geometry characteristics are recognized in a short time. All recognition results are stored in the database for further processes. A case study is used to verify the validity of the recognition method.