Plenary and Semi-Plenary Lectures
Click here to return to the list of invitees
| Title: |
| Energy conserving schemes: a key tool to design robust transient algorithms |
| Lecturer: |
| Alain Combescure |
| Abstract: |
This paper shall present the interest of energy conservation and stability control for the design of innovative transient algorithms. The paper shall show that careful control of energy permits to solve problems such as:
• coupling of sub domains with incompatible time space boundaries. This is very important for easy space meshing of complex structures and computation costs savings
• precision controlled transient computations based on automatic multigrid method and error indicator
• dynamic crack propagation using T-XFEM method: for this case it shall be shown that X-FEM tool guaranties exact energy conservation even when the crack grows. A special development of X-FEM to dynamic explicit concept shall be presented
A new extended time finite element based family of time integrator shall be presented: it will be shown that this integrator family contains the time discontinuous Galerkin as a particular case. The presentation shall be illustrated by examples and some experimental comparisons. |
 I am married with 4 children from 27 to 17 years.
I graduated in 1973 from ecole polytechnique (France).
I spent 20 years at French Atomic Commission as research engineer and team leader. I worked on simulation of buckling under complex loads (thermal and seismic) and developed general Fourier Series FE code. I was awarded T Jaeger Price in 1985 for the work on buckling and especially for the development of finite element with initial imperfection. I also worked on fracture mechanics simulations.
I then spent 6 years as professor at LMT ENS Cachan where I started to work on methods for transient non-linear computations. I specially worked on incompatible time space subdomain coupling. I also worked (bots experimentally and simulation with finite elements) on phase transformation in metals during high temperature treatment.
I have now been a professor at INSA Lyon since 4 years ago. Here I have started to work with A Gravouil on the development of X-FEM for transient crack propagation problems. |
|
