| 1. |
- Li, Ci, et al.
(författare)
-
The Poses for Equine Research Dataset (PFERD)
- 2024
-
Ingår i: Scientific Data. - : Springer Nature. - 2052-4463. ; 11:1
-
Tidskriftsartikel (refereegranskat)abstract
- Studies of quadruped animal motion help us to identify diseases, understand behavior and unravel the mechanics behind gaits in animals. The horse is likely the best-studied animal in this aspect, but data capture is challenging and time-consuming. Computer vision techniques improve animal motion extraction, but the development relies on reference datasets, which are scarce, not open-access and often provide data from only a few anatomical landmarks. Addressing this data gap, we introduce PFERD, a video and 3D marker motion dataset from horses using a full-body set-up of densely placed over 100 skin-attached markers and synchronized videos from ten camera angles. Five horses of diverse conformations provide data for various motions from basic poses (eg. walking, trotting) to advanced motions (eg. rearing, kicking). We further express the 3D motions with current techniques and a 3D parameterized model, the hSMAL model, establishing a baseline for 3D horse markerless motion capture. PFERD enables advanced biomechanical studies and provides a resource of ground truth data for the methodological development of markerless motion capture.
|
|
| 2. |
- Mellbin, Ylva, et al.
(författare)
-
A combined crystal plasticity and graph-based vertex model of dynamic recrystallization at large deformations
- 2015
-
Ingår i: Modelling and Simulation in Materials Science and Engineering. - : IOP Publishing. - 0965-0393 .- 1361-651X. ; 23:4
-
Tidskriftsartikel (refereegranskat)abstract
- A mesoscale model of microstructure evolution is formulated in the present work by combining a crystal plasticity model with a graph-based vertex algorithm. This provides a versatile formulation capable of capturing finite-strain deformations, development of texture and microstructure evolution through recrystallization. The crystal plasticity model is employed in a finite element setting and allows tracing of stored energy build-up in the polycrystal microstructure and concurrent reorientation of the crystal lattices in the grains. This influences the progression of recrystallization as nucleation occurs at sites with sufficient stored energy and since the grain boundary mobility and energy is allowed to vary with crystallographic misorientation across the boundaries. The proposed graph-based vertex model describes the topological changes to the grain microstructure and keeps track of the grain inter-connectivity. Through homogenization, the macroscopic material response is also obtained. By the proposed modeling approach, grain structure evolution at large deformations as well as texture development are captured. This is in contrast to most other models of recrystallization which are usually limited by assumptions of one or the other of these factors. In simulation examples, the model is in the present study shown to capture the salient features of dynamic recrystallization, including the effects of varying initial grain size and strain rate on the transitions between single-peak and multiple-peak oscillating flow stress behavior. Also the development of recrystallization texture and the influence of different assumptions on orientation of recrystallization nuclei are investigated. Further, recrystallization kinetics are discussed and compared to classical JMAK theory. To promote computational efficiency, the polycrystal plasticity algorithm is parallelized through a GPU implementation that was recently proposed by the authors.
|
|
| 3. |
- Mellbin, Ylva, et al.
(författare)
-
Accelerating crystal plasticity simulations using GPU multiprocessors
- 2014
-
Ingår i: International Journal for Numerical Methods in Engineering. - : Wiley. - 1097-0207 .- 0029-5981. ; 100:2, s. 111-135
-
Tidskriftsartikel (refereegranskat)abstract
- Crystal plasticity models are often used to model the deformation behavior of polycrystalline materials. One major drawback with such models is that they are computationally very demanding. Adopting the common Taylor assumption requires calculation of the response of several hundreds of individual grains to obtain the stress in a single integration point in the overlying FEM structure. However, a large part of the operations can be executed in parallel to reduce the computation time. One emerging technology for running massively parallel computations without having to rely on the availability of large computer clusters is to port the parallel parts of the calculations to a graphical processing unit (GPU). GPUs are designed to handle vast numbers of floating point operations in parallel. In the present work, different strategies for the numerical implementation of crystal plasticity are investigated as well as a number of approaches to parallelization of the program execution. It is identified that a major concern is the limited amount of memory available on the GPU. However, significant reductions in computational time – up to 100 times speedup – are achieved in the present study, and possible also on a standard desktop computer equipped with a GPU.
|
|
| 4. |
- Mellbin, Ylva, et al.
(författare)
-
An extended vertex and crystal plasticity framework for efficient multiscale modeling of polycrystalline materials
- 2017
-
Ingår i: International Journal of Solids and Structures. - : Elsevier BV. - 0020-7683. ; 125, s. 150-160
-
Tidskriftsartikel (refereegranskat)abstract
- A multiscale modeling framework for polycrystal materials is established, using a combination of an extended vertex model and a crystal plasticity formulation. The 2D vertex model is cast to incorporate a range of mesoscale processes such as grain structure evolution and the influence of second-phase particles. It is combined with a finite strain crystal plasticity formulation whereby also texture development and stored energy accumulation is traced. Computational efficiency is enhanced by GPU-parallelization. The full model captures a wide range of microstructure processes such as dynamic recrystallization, grain growth, texture evolution, anisotropic grain boundary properties as well as particle pinning effects. The macroscale material behavior is directly coupled to the evolving microstructure, for example in terms of a grain size dependent flow stress behavior. Illustrative numerical examples are provided to show the capabilities of the model. For example, the interplay between particle strengthening and grain size influence on macroscopic flow stress behavior is shown, as well as effects due to dynamic recrystallization. Special attention is given to the formulation of the vertex model as the combination of stored energy, particle pinning and anisotropic grain boundary properties give rise to intricate topological transformations which have not been previously addressed.
|
|
| 5. |
- Mellbin, Ylva
(författare)
-
Efficient crystal plasticity simulations of microstructure evolution
- 2014
-
Licentiatavhandling (övrigt vetenskapligt/konstnärligt)abstract
- One of the common tools for studying the deformation behavior of the microstructure in polycrystalline materials is crystal plasticity models. These are used to describe texture evolution and hardening due to crystallographic slip. A drawback when using crystal plasticity models is that the calculation of the slip requires solving a set of stiff differential equations for each grain in the microstructure, yielding a high computational cost. In order to reduce this cost, the program has in the present work been ported to a graphical processing unit (GPU), to utilize the capabilities for parallel performance available on the GPU. Different strategies for the numerical implementation of crystal plasticity are investigated as well as a number of approaches to parallelization of the program execution. Crystal plasticity models based on the Taylor assumption are well suited for describing the plastic deformation of polycrystal grain structures, but are not equipped to model recrystallization since the topology of the grain structure is not defined, and there is no description of inter-connectivity between grains. Therefore the crystal plasticity model is combined with a graph-based vertex algorithm in this work. This formulation is capable of capturing finite-strain deformations, development of texture and microstructure evolution through recrystallization. The polycrystal plasticity model is employed in a finite element setting and allows tracing of stored energy build-up in the microstructure and concurrent reorientation of the crystal lattices in the grains. This influences the progression of recrystallization as nucleation occurs at sites with sufficiently high stored energy gradients and since the grain boundary mobility and energy is allowed to vary with crystallographic misorientation across the boundaries. The proposed graph-based vertex model describes the topological changes to the grain microstructure and keeps track of the grain inter-connectivity. Through homogenization, the macroscopic material response is also obtained. By the proposed modeling approach, grain structure evolution at large deformations as well as texture development are captured.
|
|
| 6. |
|
|
| 7. |
|
|
| 8. |
|
|
| 9. |
|
|
| 10. |
- Mellbin, Ylva
(författare)
-
Multiscale modeling of dynamic recrystallization
- 2016
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- During thermomechanical processing of metals, changes occur in the microstructure of the material which affect its macroscopic properties. By understanding these transformations in the microstructure, it becomes possible to design the processes in a way which yields the desired properties in the finished product. For this purpose, computer simulation plays an increasingly important role.The present work is focused on developing an efficient numerical model that captures the macroscopic material behavior as well as the microstructure evolution. The main part of the thesis is made up of four papers, A-D. In paper A, different numerical solution methods for crystal plasticity are compared and implemented to run on the Graphical Processing Unit (GPU). The use of GPUs for scientific computation allows for considerable parallelism to be achieved in an ordinary desktop, or even laptop, computer, and has also been proven to be a cheap and energy efficient alternative for use in clusters. Since polycrystal plasticity is well suited for parallelization, it is shown that considerable speedup, up to a factor of 100 in some cases, can be achieved.In paper B, the crystal plasticity model is coupled with a vertex model of grain structure evolution. This provides a versatile framework which can be used to model dynamic recrystallization at large deformations. The crystal plasticity model captures hardening and texture evolution during deformation, while the vertex model describes the recrystallization process in terms of nucleation and grain growth. This model is then applied to simulations of a hot rolling process in paper C, making itpossible to study how temperature, and thereby recrystallization, affects the texture evolution during rolling, and also to study the development of inhomogeneities in the microstructure throughout the workpiece. In the final paper, D, the model from paper B is further developed such that it can also account for the effects of grain size hardening and particle pinning of migrating grain boundaries.Taken together, the four papers A-D provide a numerical simulation framework with multiscale capabilities. By taking advantage of recent developments in computer hardware and using a combination of modeling approaches, a versatile tool is established. The model is capable of describing development of crystallographic texture and dynamic recrystallization, including effects of temperature and impurities in the material. Employed in a finite element setting, the effects of the microstructure evolution on the macroscopic properties of the metal are captured, providing a powerfulconstitutive model for thermomechanical processing.
|
|
