-
Explainable artificial intelligence for mechanics: physics-informing neural networks for constitutive models
Authors:
Arnd Koeppe,
Franz Bamer,
Michael Selzer,
Britta Nestler,
Bernd Markert
Abstract:
(Artificial) neural networks have become increasingly popular in mechanics to accelerate computations with model order reduction techniques and as universal models for a wide variety of materials. However, the major disadvantage of neural networks remains: their numerous parameters are challenging to interpret and explain. Thus, neural networks are often labeled as black boxes, and their results o…
▽ More
(Artificial) neural networks have become increasingly popular in mechanics to accelerate computations with model order reduction techniques and as universal models for a wide variety of materials. However, the major disadvantage of neural networks remains: their numerous parameters are challenging to interpret and explain. Thus, neural networks are often labeled as black boxes, and their results often elude human interpretation. In mechanics, the new and active field of physics-informed neural networks attempts to mitigate this disadvantage by designing deep neural networks on the basis of mechanical knowledge. By using this a priori knowledge, deeper and more complex neural networks became feasible, since the mechanical assumptions could be explained. However, the internal reasoning and explanation of neural network parameters remain mysterious.
Complementary to the physics-informed approach, we propose a first step towards a physics-informing approach, which explains neural networks trained on mechanical data a posteriori. This novel explainable artificial intelligence approach aims at elucidating the black box of neural networks and their high-dimensional representations. Therein, the principal component analysis decorrelates the distributed representations in cell states of RNNs and allows the comparison to known and fundamental functions. The novel approach is supported by a systematic hyperparameter search strategy that identifies the best neural network architectures and training parameters. The findings of three case studies on fundamental constitutive models (hyperelasticity, elastoplasticity, and viscoelasticity) imply that the proposed strategy can help identify numerical and analytical closed-form solutions to characterize new materials.
△ Less
Submitted 12 July, 2021; v1 submitted 20 April, 2021;
originally announced April 2021.
-
Morphological stability of three-dimensional cementite rods in polycrystalline system: A phase-field analysis
Authors:
Tobias Mittnacht,
Prince Gideon Kubendran Amos,
Daniel Schneider,
Britta Nestler
Abstract:
Transformations accompanying shape-instability govern the morphological configuration and distribution of the phases in a microstructure. Owing to the influence of the microstructure on the properties of a material, the stability of three-dimensional rods in a representative polycrystalline system is extensively analysed. A multiphase-field model, which recovers the physical laws and sharp-interfa…
▽ More
Transformations accompanying shape-instability govern the morphological configuration and distribution of the phases in a microstructure. Owing to the influence of the microstructure on the properties of a material, the stability of three-dimensional rods in a representative polycrystalline system is extensively analysed. A multiphase-field model, which recovers the physical laws and sharp-interface relations, and includes grain boundary diffusion, is adopted to investigate the morphological evolution of the precipitate. Moreover, the efficiency of the numerical approach is ensured by establishing the volume-preserving chemical equilibrium through the incorporation TCFe8 (CALPHAD) data and solving phase-field evolution in the Allen-Cahn framework. The morphological evolution of the rod in the multiphase system exhibits a unique transformation mechanism which is significantly different from the evolution of an isolated finite-structure. It is realised that, in a polycrystalline arrangement, irrespective of the initial rod-size , the shape-change begins with the energy-minimising events at the triple junctions. This early transformation renders a characteristic morphology at the longitudinal ends of the structure, which introduces sufficient driving-force through the curvature-difference for the subsequent morphological changes. The continued mass transfer to the terminations, ultimately, breaks-off the rod into separate entities that are entangled in the grain boundary. With increasing aspect ratio of the rod, it is identified that the source of mass transfer, which turns into the ovulation site, shifts from the centre. This increases the number of fragmentation events and introduces satellite particle. A comprehensive understanding of the transformation kinetics and mechanism governing the morphological evolution of the rods in a polycrystalline system is rendered in this work.
△ Less
Submitted 16 June, 2020;
originally announced June 2020.
-
Experience Report: Formal Methods in Material Science
Authors:
Bernhard Beckert,
Britta Nestler,
Moritz Kiefer,
Michael Selzer,
Mattias Ulbrich
Abstract:
Increased demands in the field of scientific computation require that algorithms be more efficiently implemented. Maintaining correctness in addition to efficiency is a challenge that software engineers in the field have to face.
In this report we share our first impressions and experiences on the applicability of formal methods to such design challenges arising in the development of scientific…
▽ More
Increased demands in the field of scientific computation require that algorithms be more efficiently implemented. Maintaining correctness in addition to efficiency is a challenge that software engineers in the field have to face.
In this report we share our first impressions and experiences on the applicability of formal methods to such design challenges arising in the development of scientific computation software in the field of material science. We investigated two different algorithms, one for load distribution and one for the computation of convex hulls, and demonstrate how formal methods have been used to discover counterexamples to the correctness of the existing implementations as well as proving the correctness of a revised algorithm. The techniques employed for this include SMT solvers, and automatic and interactive verification tools.
△ Less
Submitted 14 February, 2018; v1 submitted 7 February, 2018;
originally announced February 2018.
-
A Scalable and Extensible Checkpointing Scheme for Massively Parallel Simulations
Authors:
Nils Kohl,
Johannes Hötzer,
Florian Schornbaum,
Martin Bauer,
Christian Godenschwager,
Harald Köstler,
Britta Nestler,
Ulrich Rüde
Abstract:
Realistic simulations in engineering or in the materials sciences can consume enormous computing resources and thus require the use of massively parallel supercomputers. The probability of a failure increases both with the runtime and with the number of system components. For future exascale systems it is therefore considered critical that strategies are developed to make software resilient agains…
▽ More
Realistic simulations in engineering or in the materials sciences can consume enormous computing resources and thus require the use of massively parallel supercomputers. The probability of a failure increases both with the runtime and with the number of system components. For future exascale systems it is therefore considered critical that strategies are developed to make software resilient against failures. In this article, we present a scalable, distributed, diskless, and resilient checkpointing scheme that can create and recover snapshots of a partitioned simulation domain. We demonstrate the efficiency and scalability of the checkpoint strategy for simulations with up to $40$ billion computational cells executing on more than $400$ billion floating point values. A checkpoint creation is shown to require only a few seconds and the new checkpointing scheme scales almost perfectly up to more than $260\,000$ ($2^{18}$) processes. To recover from a diskless checkpoint during runtime, we realize the recovery algorithms using ULFM MPI. The checkpointing mechanism is fully integrated in a state-of-the-art high-performance multi-physics simulation framework. We demonstrate the efficiency and robustness of the method with a realistic phase-field simulation originating in the material sciences and with a lattice Boltzmann method implementation.
△ Less
Submitted 29 January, 2018; v1 submitted 28 August, 2017;
originally announced August 2017.
-
Massively Parallel Phase-Field Simulations for Ternary Eutectic Directional Solidification
Authors:
Martin Bauer,
Johannes Hötzer,
Philipp Steinmetz,
Marcus Jainta,
Marco Berghoff,
Florian Schornbaum,
Christian Godenschwager,
Harald Köstler,
Britta Nestler,
Ulrich Rüde
Abstract:
Microstructures forming during ternary eutectic directional solidification processes have significant influence on the macroscopic mechanical properties of metal alloys. For a realistic simulation, we use the well established thermodynamically consistent phase-field method and improve it with a new grand potential formulation to couple the concentration evolution. This extension is very compute in…
▽ More
Microstructures forming during ternary eutectic directional solidification processes have significant influence on the macroscopic mechanical properties of metal alloys. For a realistic simulation, we use the well established thermodynamically consistent phase-field method and improve it with a new grand potential formulation to couple the concentration evolution. This extension is very compute intensive due to a temperature dependent diffusive concentration. We significantly extend previous simulations that have used simpler phase-field models or were performed on smaller domain sizes. The new method has been implemented within the massively parallel HPC framework waLBerla that is designed to exploit current supercomputers efficiently. We apply various optimization techniques, including buffering techniques, explicit SIMD kernel vectorization, and communication hiding. Simulations utilizing up to 262,144 cores have been run on three different supercomputing architectures and weak scalability results are shown. Additionally, a hierarchical, mesh-based data reduction strategy is developed to keep the I/O problem manageable at scale.
△ Less
Submitted 4 June, 2015;
originally announced June 2015.
