|Title||FEniCS - A Free Software Framework for Efficient Solution of Differential Equations, with Applications in Mechanics|
|Afilliation||Scientific Computing, Center for Biomedical Computing (SFF), Scientific Computing|
|Project(s)||Center for Biomedical Computing (SFF)|
|Publication Type||Talk, keynote|
|Year of Publication||2014|
|Location of Talk||Prague|
|Type of Talk||Keynote|
The talk first presents the FEniCS software for solving Partial Differential Equations (PDEs) by finite element methods. With FEniCS, variational forms can be specified with a syntax (in Python) that closely resembles the mathematical notation. The variational forms are compiled into problem-specific, highly optimized C++ code and linked with various libraries (including PETSc and Trilinos). Some simple examples of FEniCS features and applications will be shown. The talk then continues by describing a Python library built on top of FEniCS for flexible specification of systems of PDEs. The user of this library can with ease divide a system of PDEs into subsystems to be solved fully coupled. It is also very easy to experiment with various linearizations of the PDEs, in a Picard or Newton fashion. This feature is particularly important in highly nonlinear PDE problems, such as those occurring in turbulent flow models, where convergence of the sequence of linear problems often requires tailored iteration techniques. Finally some advanced applications for biomedical flows and turbulent combustion for current ongoing research projects are presented.