Skip to content

Commit

Permalink
paper update
Browse files Browse the repository at this point in the history
  • Loading branch information
thelfer committed Aug 6, 2024
1 parent 05c6003 commit d5789c1
Show file tree
Hide file tree
Showing 2 changed files with 81 additions and 19 deletions.
18 changes: 18 additions & 0 deletions docs/papers/joss/bibliography.bib
Original file line number Diff line number Diff line change
@@ -1,3 +1,21 @@
@article{bernaud_pleiades_2024,
title = {{PLEIADES}: A numerical framework dedicated to the multiphysics and multiscale nuclear fuel behavior simulation},
volume = {205},
issn = {0306-4549},
url = {https://www.sciencedirect.com/science/article/pii/S0306454924002408},
doi = {10.1016/j.anucene.2024.110577},
shorttitle = {{PLEIADES}},
abstract = {The aim of this paper is to introduce the {PLEIADES} framework offering a set of services and numerical tools to model and simulate the behavior of nucl…},
pages = {110577},
journaltitle = {Annals of Nuclear Energy},
author = {Bernaud, Stéphane and Ramière, Isabelle and Latu, Guillaume and Michel, Bruno},
urldate = {2024-08-01},
date = {2024-09-15},
langid = {american},
note = {Publisher: Pergamon},
file = {Bernaud et al. - 2024 - PLEIADES A numerical framework dedicated to the m.pdf:/home/th202608/.zotero/zotero/vzpzva46.default/zotero/storage/XPHQZJBV/Bernaud et al. - 2024 - PLEIADES A numerical framework dedicated to the m.pdf:application/pdf;Snapshot:/home/th202608/.zotero/zotero/vzpzva46.default/zotero/storage/EUKMV7YX/S0306454924002408.html:text/html},
}

@article{Helfer2015,
abstract = {The PLEIADES software environment is devoted to the thermomechanical simulation of nuclear fuel elements behaviour under irradiation. This platform is co-developed in the framework of a research cooperative program between {\'{E}}lectricit{\'{e}} de France (EDF), AREVA and the French Atomic Energy Commission (CEA). As many thermomechanical solvers are used within the platform, one of the PLEAIADES's main challenge is to propose a unified software environment for capitalisation of material knowledge coming from research and development programs on various nuclear systems. This paper introduces a tool called mfront which is basically a code generator based on C++ (Stroustrup and Eberhardt, 2004). Domain specific languages are provided which were designed to simplify the implementations of new material properties, mechanical behaviours and simple material models. mfront was recently released under the GPL open-source licence and is available on its web site: http://tfel.sourceforge.net/. The authors hope that it will prove useful for researchers and engineers, in particular in the field of solid mechanics. mfront interfaces generate code specific to each solver and language considered. In this paper, after a general overview of mfront functionalities, a particular focus is made on mechanical behaviours which are by essence more complex and may have significant impact on the numerical performances of mechanical simulations. mfront users can describe all kinds of mechanical phenomena, such as viscoplasticity, plasticity and damage, for various types of mechanical behaviour (small strain or finite strain behaviour, cohesive zone models). Performance benchmarks, performed using the Code-Aster finite element solver, show that the code generated using mfront is in most cases on par or better than the behaviour implementations written in fortran natively available in this solver. The material knowledge management strategy that was set up within the PLEIADES platform is briefly discussed. A material database named sirius proposes a rigorous material verification workflow. We illustrate the use of mfront through two case of studies: a simple FFC single crystal viscoplastic behaviour and the implementation of a recent behaviour for the fuel material which describes various phenomena: fuel cracking, plasticity and viscoplasticity.},
author = {Helfer, Thomas and Michel, Bruno and Proix, Jean-Michel and Salvo, Maxime and Sercombe, J{\'{e}}r{\^{o}}me and Casella, Michel},
Expand Down
82 changes: 63 additions & 19 deletions docs/papers/joss/paper.md
Original file line number Diff line number Diff line change
Expand Up @@ -43,14 +43,15 @@ pandoc -f markdown --bibliography=bibliography.bib --citeproc -V geometry:a4pap

# Introduction

`MFront` is an open-source code generator focused on material
knowledge [@Helfer2015;@cea_edf_mfront_2022] developed collaboratively
by the French Alternative Energies and Atomic Energy Commission (CEA) and
Électricité de France (EDF). The open-source status of `MFront` has led to
its adoption in a wide range of applications[^mfront:publications] covering
a variety of materials (ceramics, metals, concrete, woods, etc.) and physical
phenomena (viscoplasticity, plasticity, damage, etc.). It is also part of the
`PLEIADES` numerical platform, which is devoted to multi-physics nuclear fuel
`MFront` is an open-source code generator focused on material knowledge
[@Helfer2015;@cea_edf_mfront_2022] developed collaboratively by the
French Alternative Energies and Atomic Energy Commission (CEA) and
Électricité de France (EDF). The open-source status of `MFront` has led
to its adoption in a wide range of applications[^mfront:publications]
covering a variety of materials (ceramics, metals, concrete, woods,
etc.) and physical phenomena (viscoplasticity, plasticity, damage,
etc.). It is also part of the `PLEIADES` numerical platform
[@bernaud_pleiades_2024], which is devoted to multi-physics nuclear fuel
simulations and is developed by CEA and its industrial partners EDF and
Framatome.

Expand All @@ -71,7 +72,16 @@ various commercial and academic solvers such as: [`Cast3M`](http://www-cast3m.ce
[`ZSet`](http://www.zset-software.com/),
[`DIANA FEA`](https://dianafea.com/).

Additionally, the `generic` interface extends the availability of these mechanical behaviours to all solvers using the [`MFrontGenericInterfaceSupport` projet](https://thelfer.github.io/mgis/web/index.html) (MGIS) [@helfer_mfrontgenericinterfacesupport_2020], including: [`OpenGeoSys`](https://www.opengeosys.org/) [@Bilke2019], [`MFEM-MGIS`](https://thelfer.github.io/mfem-mgis/web/index.html), `MANTA`, [`mgis.fenics`](https://thelfer.github.io/mgis/web/mgis_fenics.html), [`MoFEM`](http://mofem.eng.gla.ac.uk/mofem/html), XPER, etc.
Additionally, the `generic` interface extends the availability of these
mechanical behaviours to all solvers using the
[`MFrontGenericInterfaceSupport`
projet](https://thelfer.github.io/mgis/web/index.html) (MGIS)
[@helfer_mfrontgenericinterfacesupport_2020], including:
[`OpenGeoSys`](https://www.opengeosys.org/) [@Bilke2019],
[`MFEM-MGIS`](https://thelfer.github.io/mfem-mgis/web/index.html),
`MANTA`,
[`mgis.fenics`](https://thelfer.github.io/mgis/web/mgis_fenics.html),
[`MoFEM`](http://mofem.eng.gla.ac.uk/mofem/html), XPER, etc.

The `MFrontGallery` project addresses the management of
`MFront` implementations including their compilation, unit testing and
Expand All @@ -95,14 +105,23 @@ which is discussed in Section

In summary, the project provides:

- A [`CMake`](https://cmake.org) infrastructure that can be replicated in (academic or industrial) derived projects, which allows for:
- A [`CMake`](https://cmake.org) infrastructure that can be replicated
in (academic or industrial) derived projects, which allows for:
- compiling `MFront` sources using all supported interfaces.
- executing unit tests based on `MTest` which generate `XML` result files conforming to the `JUnit` standard, compatible with continuous integration platforms such as [jenkins](https://www.jenkins.io/).
- executing unit tests based on `MTest` which generate `XML` result
files conforming to the `JUnit` standard, compatible with continuous
integration platforms such as [jenkins](https://www.jenkins.io/).
- generating documentation associated with the stored implementations.
- A documentation of best practices for handling material knowledge implemented with `MFront`, such as use of consistent unit systems, bound-aware physical quantities, consistent tangent operators, and others.
- A set of high-quality `MFront` implementations.
- A documentation of best practices for handling material knowledge
implemented with `MFront`, such as use of consistent unit systems,
bound-aware physical quantities, consistent tangent operators, and
others.
- A set of high-quality `MFront` implementations. Those implementations
are not discussed in this paper which is thus focused on the two
previous points.

This paper aims to describe the `MFrontGallery` project and is organized as follows:
This paper aims to describe the `MFrontGallery` project and is organized
as follows:

Section \ref{sec:mfm:introduction:statement_of_need} discusses the
necessity for a new approach to material knowledge management, particularly
Expand All @@ -127,7 +146,7 @@ that evolve over time due to various physical phenomena.
In `MFront`, material knowledge can be categorized as follows:

- **Material properties** are defined as functions of the current
state of the material such as the Young modulus or Poisson ratio.
state of the material such as the Young's modulus or Poisson's ratio.
- **Behaviours** describe how a material evolves and reacts locally due
to internal gradients. The material reaction is
associated with fluxes (or forces) thermodynamically conjugated to
Expand All @@ -148,7 +167,12 @@ The `MFrontGallery` project has been developed to address various
issues related to material knowledge management in safety-critical
studies:

- **Intellectual property**: Material knowledge often embodies industrial know-how that must be kept confidential. This includes highly valuable mechanical behaviours derived from extensive experimental testing in dedicated facilities. `MFrontGallery` supports creating private derived projects to protect such valuable knowledge, as detailed in Section \ref{sec:mfm:creating_derived_project}.
- **Intellectual property**: Material knowledge often embodies
industrial know-how that must be kept confidential. This includes
highly valuable mechanical behaviours derived from extensive
experimental testing in dedicated facilities. `MFrontGallery` supports
creating private derived projects to protect such valuable knowledge,
as detailed in Section \ref{sec:mfm:creating_derived_project}.
- **Portability**: safety-critical studies may involve several partners
which use different solvers for independent assessment and review.
From a single `MFront` source file, `MFrontGallery` can generate
Expand All @@ -164,7 +188,11 @@ studies:
dedicated material knowledge project based on *self-contained*
implementations, facilitate maintainability as discussed in Section
\ref{sec:mfm:introduction:implementations}.
- **Progression of the state of the art**: Safety-critical studies must reflect current scientific and engineering advancements. Thus, material knowledge, numerical methods, and software engineering need to evolve while guaranteeing the quality assurance of past, present and future simulations.
- **Progression of the state of the art**: Safety-critical studies must
reflect current scientific and engineering advancements. Thus,
material knowledge, numerical methods, and software engineering need
to evolve while guaranteeing the quality assurance of past, present
and future simulations.
- **Continuous integration and unit testing**: Each implementation includes unit tests to prevent regression during during the `MFront` development.
- **Documentation**: the project can automatically generate the documentation
associated with the various implementations. Implementations of material knowledge can be associated to essential
Expand All @@ -185,11 +213,14 @@ studies:
implementations are usually provided with solvers as ready-to-use
behaviours.

Thus, self-contained implementations describe both constitutive equations **and** material coefficients identified on a well-defined set of experiments for a particular material, while generic implementations describe only the constitutive equations.
Thus, self-contained implementations describe both constitutive
equations **and** material coefficients identified on a well-defined set
of experiments for a particular material, while generic implementations
describe only the constitutive equations.

In practice, the physical
information described by self-contained implementations may be more
complex than a set of material coefficients. For example, the Young
complex than a set of material coefficients. For example, the Young's
modulus of a material may be defined by an analytical formula and cannot
thus be reduced to a set of constants. This analytical formula shall be
part of a self-contained mechanical behaviour implementation. Of course,
Expand Down Expand Up @@ -234,6 +265,19 @@ The [`CMake`](https://cmake.org) infrastructure provides:
- functions related to documentation and website generation. Those
functions will not be described in this paper.

The following snipet shows how the `mfront_properties_library`
introduces a set of libraries describing the material properties of
Uranium dioxide:

~~~{.cmake}
mfront_properties_library(UO2
UO2_YoungModulus_Martin1989
)
~~~~
The `mfront_behaviours_library` and `mfront_models_library` are
available for behaviours and point-wise models respectively.
## Creation of a derived project
\label{sec:mfm:creating_derived_project}
Expand Down

0 comments on commit d5789c1

Please sign in to comment.