Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “compute intensive problems”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 577 records · Page 32

Attractor comparisons based on density

Recognizing a chaotic attractor can be seen as a problem in pattern recognition. Some feature vector must be extracted from the attractor and used to compare to other attractors. The field of machine learning has many methods for extracting feature vectors, including clustering methods, decision trees, support vector machines, and many others. In this work, feature vectors are created by representing the attractor as a density in phase space and creating polynomials based on this density. Density is useful in itself because it is a one dimensional function of phase space position, but representing an attractor as a density is also a way to reduce the size of a large data set before analyzing it with graph theory methods, which can be computationally intensive. The density computation in this paper is also fast to execute. In this paper, as a demonstration of the usefulness of density, the density is used directly to construct phase space polynomials for comparing attractors. Comparisons between attractors could be useful for tracking changes in an experiment when the underlying equations are too complicated for vector field modeling.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Application and Evaluation of Surrogate Models for Radiation Source Search

Surrogate models are increasingly required for applications in which first-principles simulation models are prohibitively expensive to employ for uncertainty analysis, design, or control. They can also be used to approximate models whose discontinuous derivatives preclude the use of gradient-based optimization or data assimilation algorithms. We consider the problem of inferring the 2D location and intensity of a radiation source in an urban environment using a ray-tracing model based on Boltzmann transport theory. Whereas the code implementing this model is relatively efficient, extension to 3D Monte Carlo transport simulations precludes subsequent Bayesian inference to infer source locations, which typically requires thousands to millions of simulations. Additionally, the resulting likelihood exhibits discontinuous derivatives due to the presence of buildings. To address these issues, we discuss the construction of surrogate models for optimization, Bayesian inference, and uncertainty propagation. Specifically, we consider surrogate models based on Legendre polynomials, multivariate adaptive regression splines, radial basis functions, Gaussian processes, and neural networks. We detail strategies for computing training points and discuss the merits and deficits of each method.

Cook, Jared A.↗

GraphBLAST: A High-Performance Linear Algebra-based Graph Framework on the GPU

High-performance implementations of graph algorithms are challenging to implement on new parallel hardware such as GPUs, because of three challenges: (1) difficulty of coming up with graph building blocks, (2) load imbalance on parallel hardware, and (3) graph problems having low arithmetic intensity. To address these challenges, GraphBLAS is an innovative, on-going effort by the graph analytics community to propose building blocks based in sparse linear algebra, which will allow graph algorithms to be expressed in a performant, succinct, composable and portable manner. In this paper, we examine the performance challenges of a linear algebra-based approach to building graph frameworks and describe new design principles for overcoming these bottlenecks. Among the new design principles is exploiting input sparsity, which allows users to write graph algorithms without specifying push and pull direction. Exploiting output sparsity allows users to tell the backend which values of the output in a single vectorized computation they do not want computed. Load-balancing is an important feature for balancing work amongst parallel workers. We describe the important load-balancing features for handling graphs with different characteristics. The design principles described in this paper have been implemented in "GraphBLAST", the first open-source linear algebra-based graph framework on GPU targeting high-performance computing. The results show that on a single GPU, GraphBLAST has on average at least an order of magnitude speedup over previous GraphBLAS implementations SuiteSparse and GBTL, comparable performance to the fastest GPU hardwired primitives and shared-memory graph frameworks Ligra and Gunrock, and better performance than any other GPU graph framework, while offering a simpler and more concise programming model.

97 MATHEMATICS AND COMPUTING↗

A mathematical framework for thermodynamic computing with applications to chemical reaction networks

The widespread adoption of energy-intensive computing applications has led to a growing need for energy-efficient computing approaches. Thermodynamic computing offers a promising approach for low-energy computation by leveraging the intrinsic computational capabilities of physical, chemical, or biological systems. However, the mathematical foundations of thermodynamic computing require further development to fully realize the potential energy efficiencies, as well as to assess factors like noise and operational speed. In this paper, we establish a mathematical framework for utilizing thermodynamic processes to perform fundamental operations, including addition, subtraction, multiplication, and division. We highlight the use of chemical reactions as potential computational units and explore synthetic chemical and biochemical systems as practical implementations. Additionally, we demonstrate how these principles can be applied to solving complex mathematical problems, such as ordinary differential equations (ODEs) and suggest the necessary components to implement the thermodynamic computing framework using chemical reactions based in a microfluidic device. This work enhances our understanding of thermodynamic processes for natural computing as a basis for scalable, energy-efficient computation in paradigm disruptive next-generation systems.

Cannon, William R. [Pacific Northwest National Lab↗

TURBULENT CLUSTERING OF PROTOPLANETARY DUST AND PLANETESIMAL FORMATION

We study the clustering of inertial particles in turbulent flows and discuss its applications to dust particles in protoplanetary disks. Using numerical simulations, we compute the radial distribution function (RDF), which measures the probability of finding particle pairs at given distances, and the probability density function of the particle concentration. The clustering statistics depend on the Stokes number, St, defined as the ratio of the particle friction timescale, {tau}{sub p}, to the Kolmogorov timescale in the flow. In agreement with previous studies, we find that, in the dissipation range, the clustering intensity strongly peaks at St {approx_equal} 1, and the RDF for St {approx} 1 shows a fast power-law increase toward small scales, suggesting that turbulent clustering may considerably enhance the particle collision rate. Clustering at inertial-range scales is of particular interest to the problem of planetesimal formation. At these large scales, the strongest clustering is from particles with {tau}{sub p} in the inertial range. Clustering of these particles occurs primarily around a scale where the eddy turnover time is {approx}{tau}{sub p}. We find that particles of different sizes tend to cluster at different locations, leading to flat RDFs between different particles at small scales. In the presence of multiple particle sizes, the overall clustering strength decreases as the particle size distribution broadens. We discuss particle clustering in two recent models for planetesimal formation. We argue that, in the model based on turbulent clustering of chondrule-size particles, the probability of finding strong clusters that can seed planetesimals may have been significantly overestimated. We discuss various clustering mechanisms in simulations of planetesimal formation by gravitational collapse of dense clumps of meter-size particles, in particular the contribution from turbulent clustering due to the limited numerical resolution.

79 ASTRONOMY AND ASTROPHYSICS↗

A Study of Heavy Precipitation Events in Taiwan During 10-13 August, 1994: Mesoscale Model Simulations - Part 2

Several heavy precipitation episodes occurred over Taiwan from August 10 to 13, 1994. Precipitation patterns and characteristics are quite different between the precipitation events that occurred from August 10 and I I and from August 12 and 13. In Part I (Chen et al. 2001), the environmental situation and precipitation characteristics are analyzed using the EC/TOGA data, ground-based radar data, surface rainfall patterns, surface wind data, and upper air soundings. In this study (Part II), the Penn State/NCAR Mesoscale Model (MM5) is used to study the precipitation characteristics of these heavy precipitation events. Various physical processes (schemes) developed at NASA Goddard Space Flight Center (i.e., cloud microphysics scheme, radiative transfer model, and land-soil-vegetation surface model) have recently implemented into the MM5. These physical packages are described in the paper, Two way interactive nested grids are used with horizontal resolutions of 45, 15 and 5 km. The model results indicated that Cloud physics, land surface and radiation processes generally do not change the location (horizontal distribution) of heavy precipitation. The Goddard 3-class ice scheme produced more rainfall than the 2-class scheme. The Goddard multi-broad-band radiative transfer model reduced precipitation compared to a one-broad band (emissivity) radiation model. The Goddard land-soil-vegetation surface model also reduce the rainfall compared to a simple surface model in which the surface temperature is computed from a Surface energy budget following the "force-re store" method. However, model runs including all Goddard physical processes enhanced precipitation significantly for both cases. The results from these runs are in better agreement with observations. Despite improved simulations using different physical schemes, there are still some deficiencies in the model simulations. Some potential problems are discussed. Sensitivity tests (removing either terrain or radiative processes) are performed to identify the physical processes that determine the precipitation patterns and characteristics for heavy rainfall events. These sensitivity tests indicated that terrain can play a major role in determining the exact location for both precipitation events. The terrain can also play a major role in determining the intensity of precipitation for both events. However, it has a large impact on one event but a smaller one on the other. The radiative processes are also important for determining, the precipitation patterns for one case but. not the other. The radiative processes can also effect the total rainfall for both cases to different extents.

Tao, Wei Kuo↗

Sound Diffraction Modeling of Rotorcraft Noise Around Terrain

A new computational technique, Wave Confinement (WC), is extended here to account for sound diffraction around arbitrary terrain. While diffraction around elementary scattering objects, such as a knife edge, single slit, disc, sphere, etc. has been studied for several decades, realistic environments still pose significant problems. This new technique is first validated against Sommerfeld's classical problem of diffraction due to a knife edge. This is followed by comparisons with diffraction over three-dimensional smooth obstacles, such as a disc and Gaussian hill. Finally, comparisons with flight test acoustics data measured behind a hill are also shown. Comparison between experiment and Wave Confinement prediction demonstrates that a Poisson spot occurred behind the isolated hill, resulting in significantly increased sound intensity near the center of the shadowed region.

Stephenson, James H.↗

Computational Modeling to Limit the Impact Displays and Indicator Lights Have on Habitable Volume Operational Lighting Constraints

NASA has demonstrated an interest in improving astronaut health and performance through the installment of a new lighting countermeasure on the International Space Station. The Solid State Lighting Assembly (SSLA) system is designed to positively influence astronaut health by providing a daily change to light spectrum to improve circadian entrainment. Unfortunately, existing NASA standards and requirements define ambient light level requirements for crew sleep and other tasks, yet the number of light-emitting diode (LED) indicators and displays within a habitable volume is currently uncontrolled. Because each of these light sources has its own unique spectral properties, the additive lighting environment ends up becoming something different from what was planned or researched. Restricting the use of displays and indicators is not a solution because these systems provide beneficial feedback to the crew. The research team for this grant used computer-based computational modeling and real-world lighting mockups to document the impact that light sources other than the ambient lighting system contribute to the ambient spectral lighting environment. In particular, the team was focused on understanding the impacts of long-term tasks located in front of avionics or computer displays. The team also wanted to understand options for mitigating the changes to the ambient light spectrum in the interest of maintaining the performance of a lighting countermeasure. The project utilized a variety of physical and computer-based simulations to determine direct relationships between system implementation and light spectrum. Using real-world data, computer models were built in the commercially available optics analysis software Zemax Optics Studio(c). The team also built a mockup test facility that had the same volume and configuration as one of the Zemax models. The team collected over 1200 spectral irradiance measurements, each representing a different configuration of the mockup. Analysis of the data showed a measurable impact on ambient light spectrum. This data showed that obvious design techniques exist that can be used to bind the ambient light spectrum closer to the planned spectral operating environment for the observer's eye point. The following observations should be considered when designing an operational environment that is dominated by computer displays. When more light is directed into the field of view of the observer, the greater the impact it will make on various human factors issues that depend on spectral shape and intensity. Because viewing angle has a large part to play in the amount of light flux on the crewmember's retina, beam shape, combined with light source location is an important factor for determining percent probable incident flux on the observer from any combination of light sources. Computer graphics design and display lumen output are major factors influencing the amount of spectrally intense light projected into the environment and in the viewer's direction. Use of adjustable white point display software was useful only if the predominant background color was white and if it matched the ambient light system's color. Display graphics that used a predominantly black background had the least influence on unplanned spectral energy projected into the environment. Percent reflectance makes a difference in total energy reflected back into an environment, and within certain architectural geometries, reflectance can be used to control the amount of a light spectrum that is allowed to perpetuate in the environment. Data showed that room volume and distance from significant light sources influence the total spectrum in a room. Smaller environments had a homogenizing effect on total light spectrum, whereas light from multiple sources in larger environments was less mixed. The findings indicated above should be considered when making recommendations for practice or standards for architectural systems. The ambient lighting system, surface reflectance, and display and indicator implementation all factor into the users' spectral environment. A variety of low-cost solutions exist to mitigate the impact of light from non-architectural lighting systems, and much potential for system automation and integration of display systems with the ambient environment. This team believes that proper planning can be used to avoid integration problems and also believes that human-in-the-loop evaluations, real-world test and measurement, and computer modeling can be used to determine how changes to a process, display graphics, and architecture will help maintain the planned spectral operating lighting environment.

Clark, T. A.↗

Optimization of Car Body under Constraints of Noise, Vibration, and Harshness (NVH), and Crash

To be competitive on the today's market, cars have to be as light as possible while meeting the Noise, Vibration, and Harshness (NVH) requirements and conforming to Government-man dated crash survival regulations. The latter are difficult to meet because they involve very compute-intensive, nonlinear analysis, e.g., the code RADIOSS capable of simulation of the dynamics, and the geometrical and material nonlinearities of a thin-walled car structure in crash, would require over 12 days of elapsed time for a single design of a 390K elastic degrees of freedom model, if executed on a single processor of the state-of-the-art SGI Origin2000 computer. Of course, in optimization that crash analysis would have to be invoked many times. Needless to say, that has rendered such optimization intractable until now. The car finite element model is shown. The advent of computers that comprise large numbers of concurrently operating processors has created a new environment wherein the above optimization, and other engineering problems heretofore regarded as intractable may be solved. The procedure, shown, is a piecewise approximation based method and involves using a sensitivity based Taylor series approximation model for NVH and a polynomial response surface model for Crash. In that method the NVH constraints are evaluated using a finite element code (MSC/NASTRAN) that yields the constraint values and their derivatives with respect to design variables. The crash constraints are evaluated using the explicit code RADIOSS on the Origin 2000 operating on 256 processors simultaneously to generate data for a polynomial response surface in the design variable domain. The NVH constraints and their derivatives combined with the response surface for the crash constraints form an approximation to the system analysis (surrogate analysis) that enables a cycle of multidisciplinary optimization within move limits. In the inner loop, the NVH sensitivities are recomputed to update the NVH approximation model while keeping the Crash response surface constant. In every outer loop, the Crash response surface approximation is updated, including a gradual increase in the order of the response surface and the response surface extension in the direction of the search. In this optimization task, the NVH discipline has 30 design variables while the crash discipline has 20 design variables. A subset of these design variables (10) are common to both the NVH and crash disciplines. In order to construct a linear response surface for the Crash discipline constraints, a minimum of 21 design points would have to be analyzed using the RADIOSS code. On a single processor in Origin 2000 that amount of computing would require over 9 months! In this work, these runs were carried out concurrently on the Origin 2000 using multiple processors, ranging from 8 to 16, for each crash (RADIOSS) analysis. Another figure shows the wall time required for a single RADIOSS analysis using varying number of processors, as well as provides a comparison of 2 different common data placement procedures within the allotted memories for each analysis. The initial design is an infeasible design with NVH discipline Static Torsion constraint violations of over 10%. The final optimized design is a feasible design with a weight reduction of 15 kg compared to the initial design. This work demonstrates how advanced methodology for optimization combined with the technology of concurrent processing enables applications that until now were out of reach because of very long time-to-solution.

Kodiyalam, Srinivas↗

A finite difference method of solving anisotropic scattering problems

A new method of solving radiative transfer problems is described including a comparison of its speed with that of the doubling method, and a discussion of its accuracy and suitability for computations involving variable optical properties. The method uses a discretization in angle to produce a coupled set of first-order differential equations which are integrated between discrete depth points to produce a set of recursion relations for symmetric and anti-symmetric angular sums of the radiation field at alternate depth points. The formulation given here includes depth-dependent anisotropic scattering, absorption, and internal sources, and allows arbitrary combinations of specular and non-Lambertian diffuse reflection at either or both boundaries. Numerical tests of the method show that it can return accurate emergent intensities even for large optical depths. The method is also shown to conserve flux to machine accuracy in conservative atmospheres

Barkstrom, B. R.↗

Micromechanical predictions of crack propagation and fracture energy in a single fiber boron/aluminum model composite

The axisymmetric finite element model and associated computer program developed for the analysis of crack propagation in a composite consisting of a single broken fiber in an annular sheath of matrix material was extended to include a constant displacement boundary condition during an increment of crack propagation. The constant displacement condition permits the growth of a stable crack, as opposed to the catastropic failure in an earlier version. The finite element model was refined to respond more accurately to the high stresses and steep stress gradients near the broken fiber end. The accuracy and effectiveness of the conventional constant strain axisymmetric element for crack problems was established by solving the classical problem of a penny-shaped crack in a thick cylindrical rod under axial tension. The stress intensity factors predicted by the present finite element model are compared with existing continuum results.

Adams, D. F.↗

CAT: computer aided triage improving upon the Bayes risk through ε-refusal triage rules

Manual extraction of information from electronic pathology (epath) reports to populate the Surveillance, Epidemiology, and End Result (SEER) database is labor intensive. Systematizing the data extraction automatically using machine-learning (ML) and natural language processing (NLP) is desirable to reduce the human labor required to populate the SEER database and to improve the timeliness of the data. This enables scaling up registry efficiency and collection of new data elements. To ensure the integrity, quality, and continuity of the SEER data, the misclassification error of ML and NPL algorithms needs to be negligible. Current algorithms fail to achieve the precision of human experts who can bring additional information in their assessments. Differences in registry format and the desire to develop a common information extraction platform further complicate the ML/NLP tasks. The purpose of our study is to develop triage rules to partially automate registry workflow to improve the precision of the auto-extracted information. This paper presents a mathematical framework to improve the precision of a classifier beyond that of the Bayes classifier by selectively classifying item that are most likely to be correct. This results in a triage rule that only classifies a subset of the item. We characterize the optimal triage rule and demonstrate its usefulness in the problem of classifying cancer site from electronic pathology reports to achieve a desired precision. From the mathematical formalism, we propose a heuristic estimate for triage rule based on post-processing the soft-max output from standard machine learning algorithms. We show, in test cases, that the triage rule significantly improve the classification accuracy.

60 APPLIED LIFE SCIENCES↗

Towards a Predictive Thermodynamic Model of Oxidation States of Uranium Incorporated in Fe (hydr) oxides

The theoretical research in this project has been directed toward the interpretation of core-level spectroscopies for systems relevant to the project. For the initial efforts, the focus of our theoretical simulations has been the interpretation of laboratory and synchrotron X-Ray Photoemission Spectra, XPS. In more recent efforts, an increasing emphasis has been placed on developing transparent understandings of X-Ray Adsorption Spectra, XAS . For the XAS, the principal concern is for the near-edge features, either just below or just above, an ionization limit or edge, which are described as Near-Edge X-Ray Adsorption Fine Structure, NEXAFS. In particular, a priority has involved the analysis and interpretation of XPS and NEXAFS spectra, especially of Fe and U systems, as measured by our PNNL collaborators. The overall objective of our theoretical studies is to establish connections between features of the spectra and their origin in the electronic structure of the materials. The efforts for the analysis of XPS have been reviewed in a paper by the PI, C. J. Nelin, and E. S. Ilton from PNNL on “The interpretation of XPS spectra: Insights into materials properties”, Surf. Sci. Reports, 68, 273 (2013). Two materials properties of special interest have been the degree of ionicity and the character of the covalent bonding in a range of oxides formed with transition metal, lanthanide, and actinide cations. Since the systems treated have electrons in open shells, it has been necessary to determine the energetics and the character of the angular momentum coupling of the open shell electrons. In particular, we have established methods for the treatment of the “intermediate coupling” which arises when the system is between the limit of Russell-Saunders multiplets, and the limit of j-j coupling where the spin-orbit splittings of single electrons dominate. A recent paper by the PI, and M. J. Sassi, and K. M. Rosso, (both at PNNL) “Intermediate Coupling For Core-Level Excited States: Consequences For X-Ray Absorption Spectroscopy”, J. Elec. Spectros. and Related Phenom., 200, 174 (2015) describes our first application of these methods. As well as applications to problems and materials of direct interest for our PNNL colleagues, we have pursued applications of fundamental theoretical significance for the analysis and interpretation of XPS and XAS spectra. These studies are important for the development of the fields of core-level spectroscopies as well as to advance our capabilities for applications of interest to our PNNL colleagues. An excellent example is our study of the surface core-level shifts, SCLS, for the surface and bulk atoms of an oxide that provides a new approach to understanding how the surface electronic of oxides differs from that in the bulk of the material. This work has the potential to lead to a new key to understanding the reactivity of oxide surfaces. Our theoretical studies use cluster models with finite numbers of atoms to describe the properties of condensed phases and crystals. This approach has allowed us to focus on the local atomistic, chemical interactions. For these clusters, we obtain orbitals and spinors through the solution of the Hartree-Fock, HF, and the fully relativistic Dirac HF equations. These orbitals are used to form configuration mixing wavefunctions which treat the many-body effects responsible for the open shell angular momentum coupling and for the satellites of the core-level spectra. Our efforts have been in two complementary directions. As well as the applications described above, we have placed major emphasis on the enhancement and extension of our theoretical and computational capabilities so that we can treat complex systems with a greater range of many-body effects. Noteworthy accomplishments in terms of method development and enhancement have included: (1) An improvement in our treatment of the large matrices that must be handled when many-body effects are treated. (2) Improvements and extensions of our capabilities for the calculation of the intensities of XPS and XAS transitions. And (3) ongoing development of flexible programs for the visualization of the theoretical spectra so that they can be compared with experiment. Our efforts on applications and methodology for these and related topics will continue under a sub-contract to PNNL.

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗

Stationary self-focusing of intense laser beam in cold quantum plasma using ramp density profile

By using a transient density profile, we have demonstrated stationary self-focusing of an electromagnetic Gaussian beam in cold quantum plasma. The paper is devoted to the prospects of using upward increasing ramp density profile of an inhomogeneous nonlinear medium with quantum effects in self-focusing mechanism of high intense laser beam. We have found that the upward ramp density profile in addition to quantum effects causes much higher oscillation and better focusing of laser beam in cold quantum plasma in comparison to that in the classical relativistic case. Our computational results reveal the importance and influence of formation of electron density profiles in enhancing laser self-focusing.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Simulation requirements for the Large Deployable Reflector (LDR)

Simulation tools for the large deployable reflector (LDR) are discussed. These tools are often the transfer function variety equations. However, transfer functions are inadequate to represent time-varying systems for multiple control systems with overlapping bandwidths characterized by multi-input, multi-output features. Frequency domain approaches are the useful design tools, but a full-up simulation is needed. Because of the need for a dedicated computer for high frequency multi degree of freedom components encountered, non-real time smulation is preferred. Large numerical analysis software programs are useful only to receive inputs and provide output to the next block, and should be kept out of the direct loop of simulation. The following blocks make up the simulation. The thermal model block is a classical heat transfer program. It is a non-steady state program. The quasistatic block deals with problems associated with rigid body control of reflector segments. The steady state block assembles data into equations of motion and dynamics. A differential raytrace is obtained to establish a change in wave aberrations. The observation scene is described. The focal plane module converts the photon intensity impinging on it into electron streams or into permanent film records.

Soosaar, K.↗

Mapping a battlefield simulation onto message-passing parallel architectures

Perhaps the most critical problem in distributed simulation is that of mapping: without an effective mapping of workload to processors the speedup potential of parallel processing cannot be realized. Mapping a simulation onto a message-passing architecture is especially difficult when the computational workload dynamically changes as a function of time and space; this is exactly the situation faced by battlefield simulations. This paper studies an approach where the simulated battlefield domain is first partitioned into many regions of equal size; typically there are more regions than processors. The regions are then assigned to processors; a processor is responsible for performing all simulation activity associated with the regions. The assignment algorithm is quite simple and attempts to balance load by exploiting locality of workload intensity. The performance of this technique is studied on a simple battlefield simulation implemented on the Flex/32 multiprocessor. Measurements show that the proposed method achieves reasonable processor efficiencies. Furthermore, the method shows promise for use in dynamic remapping of the simulation.

Nicol, David M.↗

Turbulent Premixed Flame Propagation in Microgravity

A facility in which turbulent Couette flow could be generated in a microgravity environment was designed and built. To fit into the NASA Lewis drop tower the device had to be very compact. This means that edge effects and flow re-circulation were expected to affect the flow. The flow was thoroughly investigated using LDV and was found to be largely two dimensional away from the edges with constant turbulence intensities in the core. Slight flow asymmetries are introduced by the non symmetric re-circulation of the fluid outside the test region. Belt flutter problems were remedied by adding a pair of guide plates to the belt. In general, the flow field was found to be quite similar to previously investigated Couette flows. However, turbulence levels and associated shear stresses were higher. This is probably due to the confined re-circulation zone reintroducing turbulence into the test section. An estimate of the length scales in the flow showed that the measurements were able to resolve nearly all the length scales of interest. Using a new LES method for subgrid combustion it has been demonstrated that the new procedure is computational feasible even on workstation type environment. It is found that this model is capable of capturing the propagation of the premixed names by resolving the flame in the LES grid within 2-3 grid points. In contrast, conventional LES results in numerical smearing of the flame and completely inaccurate estimate of the turbulent propagation speed. Preliminary study suggests that there is observable effect of buoyancy in the 1g environment suggesting the need for microgravity experiments of the upcoming experimental combustion studies. With the cold flow properties characterized, an identical hot flow facility is under construction. It is assumed that the turbulence properties ahead of the flame in this new device will closely match the results obtained here. This is required since the hot facility will not enable LDV measurements. The reacting flow facility is also being constructed with planned drop tower experiments in mind. Therefore, issues related to safety and structural integrity are being take into account. Further development of the numerical model will also be carried out to include finite-rate kinetics for representative premixed cases. More detail analysis of the flame structure and propagation nature will be investigated. Simulations will also be compared to the flame properties observed in the experiments.

Menon, S.↗

A comparative study of shear band tracking strategies in three-dimensional finite elements with embedded weak discontinuities

We present a computational framework for the treatment of shear localization in metallic materials under dynamic loading, based on the integration of a shear band tracking strategy into an explicit 3D finite element formulation with embedded weak discontinuities. Within this computational framework, an embedded shear band’s mid-surface is represented by an iso-surface of a level set function, which is obtained by solving a heatconduction type boundary value problem (BVP). The solution of this BVP is carried out either globally over the entire problem domain, or locally at the level of individual elements in the mesh. In this paper, we present a detailed comparison of these global and local algorithmic implementations of the shear band tracking strategy. Numerical results obtained using these two implementations, and using a simplified formulation without shear band tracking, are presented and compared. Moreover, we compare the computational efficiency and parallel scaling performance of the local and global implementations. This comparative study shows that both implementations can simulate the initiation of two independent shear bands and their propagation past each other without merging, whereas only the global implementation can successfully simulate the merging of two branches of a single shear band. This study also confirms that the global implementation is more computationally intensive, since it requires a global system of linear equations associated with the level set BVP to be solved at each time step. Both implementations exhibit very good scalability in domain decomposition-based parallel simulations.

36 MATERIALS SCIENCE↗