Development and performance evaluation of a web-based feature extraction and recognition system for sheet metal bending process planning operations

ABSTRACT

Sheet metal bending manufacturing companies require changeable and adaptable process planning systems to shorten the production cycle time and reduce operations costs. This is due to globalisation and the rapid change in market demands for sheet metal products. In light of this, this paper proposes a web-based feature extraction and recognition system. The system would ensure automated planning of various processes used by a bending machine to produce varieties of sheet metal products. The algorithms were implemented using C++ to produce the geometric and feature models used to extract, and recognise the bending features in various CAD files acquired from literature. Five (5) CAD files of various sheet metals were utilised to test the functionality of this system. The results from the feature recognition process have proven to be precisely what the user has designed and saved in the model file. Bend radius and bend angle. Finally, the developed system is able to transform a STEP file into a feature model and display the 3D CAD model within the least time. It is a cost-efficient standalone system that provides data storage, allows collaboration and data sharing. can be used anywhere where there is internet access.

KEYWORDS:

• web based systems
• Feature extraction
• feature recognition
• step
• process planning

Acknowledgments

This work was supported and fully funded by the Schlumberger Faculty for The Future. Authors are pleased to acknowledge the research team members of the Industrial engineering department, in the Faculty of Engineering and Built Enviroment at the Tshwane University of Technology (TUT), Pretoria, for the assistance and support in writing the paper and Gibela The Department of Industrial and Manufacturing Engineering in the Faculty of Engineering at the National University of Sceince and Technology (NUST) also supported the research.

Disclosure statement

No potential conflict of interest was reported by the authors.