Objective
The objective of DEEPWELD is to develop a multi-physics multi-scale numerical tool for simulating the FrictionStir Welding process, that will be able to obtain accurate predictions of residual stresses, weld properties andtool loads. The new tool will be a large step forward compared to current thermo-mechanical solutions, indeedthe latter depend on well calibrated heat fluxes used as energy input in the model. These heat fluxes must beobtained by measurements of data during experiments. This prevents op timisation of the process parametersand real usage of the FSW simulation system in a design environment for predictive simulation of the behaviourof parts.A novel approach will be followed in DEEPWELD. A detailed simulation of the material flow around th e tool willbe coupled to a simulation of the complete welding process. The software developments forming the core ofDEEPWELD will be focussed on: a material flow solver capable of simulating the material flow around the tool; coupling of the material flo w solver to existing but modified industrial FE codes; this will allow the use ofall existing features of the industrial codes to handle industrial applications and ease exploitation of thenew tool in the industrial environment. since the temperature wil l strongly affect the rheology of the material, a material model (metallurgy)capable of dealing with the large variations in material properties throughout the weld region will bedeveloped and implemented; furthermore, an appropriate friction law which d escribes the behaviour at the interface between tool andweld material needs to be implemented.DEEPWELD also involves a strong experimental part. Experiments will be conducted at two levels to obtain onone hand the required input data for the simulation such as accurate reologhical material properties, improvedmetallurgy models, friction models, material flow visualization and on the other hand detailed #
Fields of science
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
- natural sciencesphysical sciencesopticsmicroscopyelectron microscopy
- natural sciencescomputer and information sciencescomputational sciencemultiphysics
- engineering and technologymaterials engineeringmetallurgy
- natural sciencescomputer and information sciencessoftwaresoftware applicationssimulation software
- social scienceslaw
Topic(s)
Call for proposal
FP6-2003-AERO-1
See other projects for this call
Funding Scheme
STIP - Specific Targeted Innovation ProjectCoordinator
GOSSELIES
Belgium