WeldVue: Optimised welding in high value industries
Background
One of the most critical processes employed in the manufacturing industry is welding. It is applied to a variety of materials and products in different industrial sectors (e.g. steel fabrication, automotive, aerospace, construction). Therefore, the condition of welding has direct impact on the quality of final products beyond a good design and high-quality materials. Welds should have the proper penetration and the correct welding process and technique must be used for the particular alloy, position, and overall condition.
Currently, the growing use of robots in welding processes, particularly in the automotive sector, is improving and accelerating production. However, quality monitoring is still rather manual: testing of parts happens after different stages of manufacturing and in some cases with destructive techniques (i.e., welded samples are cut to be inspected).
Therefore, the technological challenge is to provide a solution enabling intelligent optimisation and monitoring of welding processes in serial production/welding of metal parts for manufacturing sectors.
Objectives
The objectives of the WeldVue project are to:
- Implement an advanced AI-based model for automotive parts manufacturing processes, optimisation and reconfiguration (resulting in a process reconfiguration time reduction of 15%);
- Implement an operational pilot line with sensors and quality inspection NDT systems for process and product quality monitoring and inspection, with probability of defect detection > 99%);
- Improve quality control by generating 50% less scrap and waste and near-zero defect parts;
- Increase process efficiency by decreasing downtime by 15%
Benefits
The WeldVue solution will allow;
- Applicability to any type of welding technique;
- Fast resolution of welding parameters from weld geometry specifications (less than a second);
- Automatic identification of possible welding defects by analysing the process parameters through monitoring weld features, benchmarking and product requirements
- Effective provision of welding procedure specifications for repair;
- Possibility of testing the series of optimal welding parameters against cost models and manufacturability.
These characteristics will enable manufacturing of near-zero defect parts, reducing scrap waste, and fast delivery of quality products to market, meeting customers’ quality requirements and needs.
Brunel Innovation Centre's Role
Brunel Innovation Centre will validate the initial model and develop algorithms to be used in WeldVue, linking with industry partners and Tier 1 manufacturers in order to turn the developments into market-based solutions. In addition to the design and validation of the model and algorithms, BIC will assist in ensuring technical robustness of the platform.
Project Partners
- STL Tech Ltd
- Ether NDE Ltd
- Teknopar Endustriyel Otomasyon San. A.S.
- Coskunoz Kalip Makina San Ve Tic. A.S.
- TWI Ltd
- Brunel University London