This project aims to mathematically model multiphase flow in the steel continuous casting in order to gain increased understanding and practical insights to improve this important commercial process. Specifically, Large Eddy Simulations of turbulent fluid flow are conducted to investigate the dynamic motion of argon bubbles in the caster with different casting conditions such as electro-magnetic braking (EMBr), in order to minimize inclusion entrapment. The present work quantifies how the oscillations of the shape and velocity of the rising bubbles can be damped with the application of a static external magnetic field.

Micro- and nanoscale tubular structures can be formed by strain-induced self-rolled-up nanomembranes. Precision engineering of the shape and dimension determines the performance of devices based on this platform for electronic, optical, and biological applications. A transient quasi-static finite element method (FEM) with moving boundary conditions is proposed as a general approach to design diverse types of three-dimensional (3D) rolled-up geometries. This method captures the dynamic release process of membranes through etching driven by mismatch strain and accurately predicts the final dimensions of rolled-up structures. Guided by the FEM modeling, experimental demonstration using silicon nitride membranes was achieved with unprecedented precision including controlling fractional turns of a rolled-up membrane, anisotropic rolling to form helical structures, and local stress control for 3D hierarchical architectures.

The Blue Waters system at the National Center for Supercomputing Applications (NCSA) is the largest GPU accelerated system in the NSF's portfolio with greater than (>) 4200 Nvidia K20x accelerators and greater than (>) 22500 compute nodes overall. Using the accelerator nodes effectively is paramount to the system's success as they represent approximately 1/7 of system peak performance. As an XSEDE level 2 service provider, the system is also available to education allocations proposed by XSEDE educators and trainers. The training staff working at Pittsburgh Supercomputing Center (PSC) along with their XSEDE and Nvidia partners have offered multiple OpenACC workshops since 2012. The most recent workshop was conducted on Blue Waters hosting the hands-on sessions and it was very successful. As a direct result of working with PSC on these workshop, NCSA researchers have been able to obtain significant speedups on real-world algorithms using OpenACC in the Cray environment. In this work we will look at two key kernel codes (3D FFT kernel, Laplace 2D MPI benchmark) and the path to obtaining the observed performance gains.

Alya is the BSC in-house HPC-based multi-physics simulation code. It is designed from scratch to run efficiently in parallel supercomputers, solving coupled problems. The target domain is engineering, with all its particular features: complex geome- tries and unstructured meshes, coupled multi-physics with exotic coupling schemes and Physical models, ill-posed problems, flexibility needs for rapidly including new models, etc. Since its conception in 2004, Alya has shown scaling behaviour in an increasing number of cores. In this paper, we present its performance up to 100.000 cores in Blue Waters, the NCSA supercomputer. The selected tests are representative of the engineering world, all the problematic features included: incompressible flow in a hu- man respiratory system, low Mach combustion problem in a kiln furnace and coupled electro-mechanical problem in a heart. We show scalability plots for all cases, discussing all the aspects of such kind of simulations, including solvers convergence.

Alya is the BSC in-house HPC-based multi-physics simulation code. It is designed from scratch to run efficiently in parallel supercomputers, solving coupled problems. The target domain is engineering, with all its particular features: complex geometries and unstructured meshes, coupled multi-physics with exotic coupling schemes and Physical models, ill-posed problems, flexibility needs for rapidly including new models, etc. Since its conception in 2004, Alya has shown scaling behaviour in an increasing number of cores. In this paper, we present its performance up to 100.000 cores in Blue Waters, the NCSA supercomputer. The selected tests are representative of the engineering world, all the problematic features included: incompressible flow in a human respiratory system, low Mach combustion problem in a kiln furnace and coupled electro-mechanical problem in a heart. We show scalability plots for all cases, discussing all the aspects of such kind of simulations, including solvers convergence.

Thermo-mechanical steel solidification models, based on highly nonlinear elastic visco-plastic constitutive laws in solid and featuring efficient and robust local implicit integration scheme, are coupled with cfd turbulent calculations in the liquid pool via enhanced latent heat method. The new multi-physics model of metal solidification is applied to calculate temperature, stress, and deformation of solidifying shell in a commercial caster with real geometry. INTRODUCTION: Many manufacturing and fabrication processes such as foundry shape casting, continuous casting and welding have common solidification phenomena. One of the most important and complex of these is continuous casting, which produces 90% of steel today. Even though the process is constantly improving, there is still a significant need to minimize defects and to maximize quality and efficiency. The difficulty of plant experiments under harsh operating conditions makes computational modeling an important tool in the design and optimization of these processes. Increased computing power and better numerical methods have enabled researchers to develop better models of many different aspects of these processes. Coupling together the different models of heat transfer, solidification distortion, stress generation and turbulent fluid flow to make accurate predictions of the entire real processes remains a challenge. PROCEDURES, RESULTS AND DISCUSSION: Inertial effects are negligible in solidification problems, so using the static mechanical equilibrium as the governing equation is appropriate. ( ) 0 x b ∇ ⋅ σ + = (1) The rate decomposition of total strain in this elastic-viscoplastic model is given by: th ie el ε ε ε ε     + + = (2) where el ie th , ,    ε ε ε are the elastic, inelastic, and thermal strain rate tensors respectively. Viscoplastic strain includes both strain-rate independent plasticity and time dependant creep. Creep is significant at the high temperatures of the solidification processes and is indistinguishable from plastic strain [Kozlowski 1992] proposed a unified formulation with the following functional form to define inelastic strain.

. Alya is a multi-physics simulation code developed at Barcelona Supercomputing Center (BSC). From its inception Alya code is designed using advanced High Performance Computing programming techniques to solve coupled problems on supercomputers efficiently. The target domain is engineering, with all its particular features: complex geometries and unstructured meshes, coupled multi-physics with exotic coupling schemes and physical models, ill-posed problems, flexibility needs for rapidly including new models, etc. Since its beginnings in 2004, Alya has scaled well in an increasing number of processors when solving single-physics problems such as fluid mechanics, solid mechanics, acoustics, etc. Over time, we have made a concerted effort to maintain and even improve scalability for multi-physics problems. This poses challenges on multiple fronts, including: numerical models, parallel implementation, physical coupling models, algorithms and solution schemes, meshing process, etc. In this paper, we introduce Alya’s main features and focus particularly on its solvers. We present Alya’s performance up to 100.000 processors in Blue Waters, the NCSA supercomputer with selected multi-physics tests that are representative of the engineering world. The tests are incompressible flow in a human respiratory system, low Mach combustion problem in a kiln furnace, and coupled electro-mechanical contraction of the heart. We show scalability plots for all cases and discuss all aspects of such simulations, including solver convergence.

AbstractThe complex structure and mechanics of elastoplastic functionally graded materials (FGM) is studied from the standpoint of fractal geometry. First, upon introducing the fineness as the number of grains of either phase across the FGM, the two-phase FGM is characterized using fractals, and an interfacial fractal dimension is estimated for varying degrees of fineness. A variation in local fractal dimension is considered across or along the FGM domain, and it is characterized by Fourier series and Beta function fits. Assuming the FGM is made of locally isotropic Titanium (Ti) and Titanium Monoboride (TiB), pure shear tests are simulated using ABAQUS for fineness levels of 50, 100, and 200 under the uniform kinematic boundary condition (UKBC) and the uniform static boundary condition (USBC). The material response observed under these BCs shows high sensitivity of these systems to loading conditions. Furthermore, plastic evolution of Ti grains, assuming isotropic plastic hardening, displays a fractal, p...

A mechanisms-based fracture model applicable to a broad class of cemented aggregates and, among them, plastic-bonded explosive (PBX) composites, is presented. The model is calibrated for PBX 9502 using the available experimental data under uniaxial compression and tension gathered at various strain rates and temperatures. We show that the model correctly captures inelastic stress-strain responses prior to the load peak and it predicts the post-critical macro-fracture processes, which result from the growth and coalescence of micro-cracks. In our approach, the fracture zone is embedded into elastic matrix and effectively weakens the material's strength along the plane of the dominant fracture.

A coupled thermo-mechanical model of solidifying shell (Koric, 2006, 2011), (Hibbeler, 2009) in Abaqus/Standard is combined with turbulent fluid flow in the liquid pool and thermal distortion of the mold to create an accurate multiphysics model of steel continuous casting. The new model is applied to calculate temperature stress and deformation in a commercial beam blank caster with complex geometry. Results from the complete system compare favorably with plant measurements of shell thickness.

Elastic–plastic transitions were investigated in three-dimensional (3D) macroscopically homogeneous materials, with microscale randomness in constitutive properties, subjected to monotonically increasing, macroscopically uniform loadings. The materials are cubic-shaped domains (of up to 100 × 100 × 100 grains), each grain being cubic-shaped, homogeneous, isotropic and exhibiting elastic–plastic hardening with a J 2 flow rule. The spatial assignment of the grains’ elastic moduli and/or plastic properties is a strict-white-noise random field. Using massively parallel simulations, we find the set of plastic grains to grow in a partially space-filling fractal pattern with the fractal dimension reaching 3, whereby the sharp kink in the stress–strain curve of individual grains is replaced by a smooth transition in the macroscopically effective stress–strain curve. The randomness in material yield limits is found to have a stronger effect than that in elastic moduli. The elastic–plastic transitions in 3D simulations are observed to progress faster than those in 2D models. By analogy to the scaling analysis of phase transitions in condensed matter physics, we recognize the fully plastic state as a critical point and, upon defining three order parameters (the ‘reduced von-Mises stress’, ‘reduced plastic volume fraction’ and ‘reduced fractal dimension’), three scaling functions are introduced to unify the responses of different materials. The critical exponents are universal regardless of the randomness in various constitutive properties and their random noise levels.

Separate three-dimensional models of thermo-mechanical behavior of the solidifying shell, turbulent fluid flow in the liquid pool, and thermal distortion of the mold are combined to create an accurate multiphysics model of metal solidification at the continuum level. The new system is applied to simulate continuous casting of steel in a commercial beam-blank caster with complex geometry. A transient coupled elastic-viscoplastic model computes temperature and stress in a transverse slice through the mushy and solid regions of the solidifying metal. This Lagrangian model features an efficient numerical procedure to integrate the constitutive equations of the deltaferrite and austenite phases of solidifying steel shell using a fixed-grid finite-element approach. The Navier-Stokes equations are solved in the liquid pool using the standard K-e turbulent flow model with standard wall laws at the mushy zone edges that define the domain boundaries. The superheat delivered to the shell is incorporated into the thermal-mechanical model of the shell using a new enhanced latent heat method. Temperature and thermal distortion modeling of the complete complex-shaped mold includes the tapered copper plates, water cooling slots, backing plates, and nonlinear contact between the different components. Heat transfer across the interfacial gaps between the shell and the mold is fully coupled with the stress model to include the effect of shell shrinkage and gap formation on lowering the heat flux. The model is validated by comparison with analytical solutions of benchmark problems of conduction with phase change, and thermal stress in an unconstrained solidifying plate. Finally, results from the complete system are shown to compare favorably with plant measurements of shell thickness.

Separate three-dimensional (3-D) models of thermomechanical behavior of the solidifying shell, turbulent fluid flow in the liquid pool, and thermal distortion of the mold are combined to create an accurate multiphysics model of metal solidification at the continuum level. The new system is applied to simulate continuous casting of steel in a commercial beam-blank caster with complex geometry. A transient coupled elastic-viscoplastic model [1] computes temperature and stress in a transverse slice through the mushy and solid regions of the solidifying metal. This Lagrangian model features an efficient numerical procedure to integrate the constitutive equations of the delta-ferrite and austenite phases of solidifying steel shell using a fixed-grid finite-element approach. The Navier-Stokes equations are solved in the liquid pool using the standard K–ϵ turbulent flow model with standard wall laws at the mushy zone edges that define the domain boundaries. The superheat delivered to the shell is incorporated into the thermalmechanical model of the shell using the enhanced latent heat method [2]. Temperature and thermal distortion modeling of the complete complex-shaped mold includes the tapered copper plates, water cooling slots, backing plates, and nonlinear contact between the different components. Heat transfer across the interfacial gaps between the shell and the mold is fully coupled with the stress model to include the effect of shell shrinkage and gap formation on lowering the heat flux. The model is validated by comparison with analytical solutions of benchmark problems of conduction with phase change [3], and thermal stress in an unconstrained solidifying plate [4]. Finally, results from the complete system compare favorably with plant measurements of shell thickness.

An efficient new method has been developed to incorporate the effects of heat transfer in a liquid pool into models of heat conduction with solidification. The procedure has been added into the commercial package Abaqus [1] as a user-defined subroutine (UMATHT). Computational results of fluid flow and heat transfer in a liquid domain can be characterized by the heat flux crossing the boundary representing the solidification front, or liquidus temperature. This “superheat flux” can be incorporated into an uncoupled transient simulation of heat transfer phenomena in the mushy and solid regions by enhancing latent heat. The new method has been validated and compared to semianalytical solutions and two other numerical methods on simple test problems: two-dimensional, steady-state ledge formation in cryolite in aluminum extraction cells, and shell thinning in continuous casting of steel. Its real power, however, is for multiphysics simulations involving complex phenomena, such as solidification stress analysis with nonlinear constitutive equations. Including the superheat flux from a thermal-fluid flow simulation of the liquid pool into the latent heat provides a very efficient and robust method for incorporating the effects of fluid flow in the liquid pool into thermal-stress problems, especially for transient problems.