26 May 2022 in ACM Transactions on Mathematical Software
We develop a method for generating degree-of-freedom maps for arbitrary order Ciarlet-type finite element spaces for any cell shape. The approach is based on the composition of permutations and transformations by cell sub-entity. Current approaches to generating degree-of-freedom maps for arbitrary order problems typically rely on a consistent orientation of cell entities that permits the definition of a common local coordinate system on shared edges and faces. However, while orientation of a mesh is straightforward for simplex cells and is a local operation, it is not a strictly local operation for quadrilateral cells and, in the case of hexahedral cells, not all meshes are orientable. The permutation and transformation approach is developed for a range of element types, including arbitrary degree Lagrange, serendipity, and divergence- and curl-conforming elements, and for a range of cell shapes. The approach is local and can be applied to cells of any shape, including general polytopes and meshes with mixed cell types. A number of examples are presented and the developed approach has been implemented in open-source libraries.
18 June 2019 in Journal of Open Source Software
Adjoint models play an important role in scientific computing. They enable for instance sensitivity and stability analysis, goal-oriented mesh adaptivity and optimisation. However, the derivation and implementation of adjoint models is challenging, especially for models governed by non-linear or time-dependent partial differential equations (PDEs). In (Farrell, Ham, Funke, & Rognes, 2013), the authors proposed to automatically derive adjoint models through high-level algorithmic differentiation, where the forward model is considered as a sequence of variational problems. The implementation, named dolfin-adjoint, automatically and robustly derives adjoint models for models written in the finite element software FEniCS (Logg, Mardal, Wells, & others, 2012). However, the assumption that the model consists of a sequence of variational problems can be limiting. For instance when considering Dirichlet boundary conditions that are not explicitly stated in the variational formulation, when considering complex functionals that cannot be represented as an integral, or when coupling FEniCS to other non-PDE models.
1 January 2019 in SIAM Journal on Scientific Computing
An important step in shape optimization with partial differential equation constraints is to adapt the geometry during each optimization iteration. Common strategies are to employ mesh deformation ...
1 August 2020 in Computer Methods in Applied Mechanics and Engineering
30 April 2024 in Journal of Open Source Software
We introduce ADIOS4DOLFINx, a checkpointing framework for the latest version of the FEniCS project, known as DOLFINx. DOLFINx is a general framework for solving partial differential equations using the finite element method. The input to simulations using the finite element method is the computational domain (mesh), mesh markers, initial conditions, and boundary conditions. To be able to restart a simulation at any point, one has to have the capability to read in all of the aforementioned variables. The adios4dolfinx package implements all of these operations, using the Message Passing Interface (MPI) for communication across multiple processes and ADIOS2 for writing/reading data to/from file. In particular, the functionality of adios4dolfinx includes N-to-M-checkpointing, which means that one can store a result of a simulation that was generated with N number of processes, and read it into a program running on M processes.
19 October 2023 in Journal of Open Source Software
Recent advances in microscopy and 3D reconstruction methods have allowed for characterization of cellular morphology in unprecedented detail, including the irregular geometries of intracellular subcompartments such as membrane-bound organelles. These geometries are now compatible with predictive modeling of cellular function. Biological cells respond to stimuli through sequences of chemical reactions generally referred to as cell signaling pathways. The propagation and reaction of chemical substances in cell signaling pathways can be represented by coupled nonlinear systems of reaction-transport equations. These reaction pathways include numerous chemical species that react across boundaries or interfaces (e.g., the cell membrane and membranes of organelles within the cell) and domains (e.g., the bulk cell volume and the interior of organelles). Such systems of multi-dimensional partial differential equations (PDEs) are notoriously difficult to solve because of their high dimensionality, non-linearities, strong coupling, stiffness, and potential instabilities. In this work, we describe Spatial Modeling Algorithms for Reactions and Transport (SMART), a finite-element-based simulation package for model specification and numerical simulation of spatially-varying reaction-transport processes. SMART is based on the FEniCS finite element library, provides a symbolic representation framework for specifying reaction pathways, and supports geometries in 2D and 3D including large and irregular cell geometries obtained from modern ultrastructural characterization methods.