Research engineer

Université de technologie de Compiègne is recruiting a contractual research engineer in the framework of the work of the joint laboratory FuseMétal within the department of mechanical engineering - laboratory Roberval.

The FuseMetal laboratory is a joint laboratory between the UTC, the CNRS and the ArcelorMittal company (research center of Montataire in the Oise), created in April 2019 and supported by European Funds for Regional Development (FEDER) through the Hauts-de-France region. In an industrial context of strong development of very high strength steels, which represents one of the keys to reducing vehicle masses and CO2 emissions, the FuseMetal laboratory concretizes the scientific and technological cooperation of these partners, to optimize the processes of assembly by welding of high yield strength steels, as well as their forming with the introduction of additive manufacturing for the manufacture of tools. This "hors-murs" laboratory has about 30 members, including researchers and technical staff from UTC, CNRS and ArcelorMittal, PhD students and research engineers specially recruited for the needs of the joint laboratory.

Place of work

Compiègne

Type of contract and anticipated starting date

Fixed term 12 month contract : starting ASAP

Experience

This position would represent an initial professional experience for someone who has recently obtained their PhD in a computational mechanics framework with an experimental mechanics sensibility.

Gross monthly salary

according to experience

Workload

1 607 hours per annum

Scientific framework

Various industrial processes use lasers to manufacture or assemble metal parts. Among them, two processes stand out: additive manufacturing and sheet metal welding. Laser additive manufacturing is one of the most widely used industrial additive manufacturing processes for steel parts, due to its relatively good maturity and its ability to produce very complex parts. The laser welding of blanks allows the shaping of parts optimized in terms of thickness and material because of the assembly possibilities it offers.

For these processes based on the fusion of the material, it is necessary to understand the physical phenomena involved to guarantee the quality and the required properties of the parts produced.

The material state of the parts produced is strongly affected by the laser-material interaction and the process parameters such as the power of the laser source and its stability and, in the case of additive manufacturing, the scan path and speed, the gas environment, the thickness

of the material deposited at each layer. The choice of process parameters influences the metallurgy as well as the mechanical behavior up to rupture. It can also lead to defects in additive manufacturing: high roughness and surface defects, cracks, delamination between layers, micro-structural heterogeneities.

The simulation of processes, at the scale of the melted zone, of additive manufacturing or of welding implies the resolution of strongly coupled multiphysical problems as well as very varied time and space scales. The monolithic simulation of these processes, carried out at the finest time and space scales in order to guarantee the predictivity of the results obtained, leads to very important and even prohibitive calculation times. In order to limit the computation time, simplifying assumptions can be made on the physical phenomena at stake as well as on the levels of coupling between the physics. Nevertheless, these simplifications affect the predictivity of the results and their exploitation in the process of optimization of the process parameters. It is proposed to explore an alternative and complementary track, based on the development of numerical tools of field transfer and model reduction allowing a modeling with the right space-time discretization of each physics.

Mission

In this context, the person recruited will develop a computational strategy that will allow, for multiphysics problems, to control the accuracy and predictivity of numerical results while ensuring computation times compatible with the use of such tools in an industrial context.

More precisely, the goal is to develop a generic tool allowing to couple different simulated physics, each with the temporal and spatial discretization most adapted to the characteristic times and dimensions of the phenomena at stake. The challenge is then to develop coupling tools between the physics that allow to guarantee and transfer the physically relevant quantities between numerical models. In a complementary way, it can be considered to replace certain physics (or "blocks of physics") by reduced models.

The objective of the work is to demonstrate the relevance of such an approach based on a case study well mastered and representative of the problem. The developments will be done within Comsol Multiphysics by potentially using the JAVA language or by coupling Comsol Multiphysics with other programming environments (Matlab, Python, ...) with the perspective of an application to the case of welding processes and/or additive manufacturing.

Principal activities

- Develop a computational strategy for multiphysics problems

- Ensure the numerical implementation of the developed method

- Propose tools for transferring fields between numerical models as well as tools for model reduction

- Validate the developments from complete simulations on a case study

- Write progress reports and promote the work through communications and publications

- Participate in the various meetings of the project and of the FuseMétal joint laboratory

- Present the results of the project.