The recent interests in developing multiscale model-based simulation procedures have brought about the challenging tasks of bridging different spatial and temporal scales within a unified framework. However, the research focus has been on the scale effect in the spatial domain with the loading rate being assumed to be quasi-static. Although material properties are rate-dependent in nature, little has been done in understanding combined loading rate and specimen size effects on the material properties at different scales. In addition, the length and time scales that can be probed by the molecular level simulations are still fairly limited due to the limitation of computational capability. Based on the experimental and computational capabilities available, therefore, an attempt is made in this report to formulate a hyper-surface in both spatial and temporal domains to predict combined size and rate effects on the mechanical properties of engineering materials. To demonstrate the features of the proposed hyper-surface, tungsten specimens of various sizes under various loading rates are considered with a focus on the uniaxial loading path. The mechanical responses of tungsten specimens under other loading paths are also explored to better understand the size effect. It appears from the preliminary results that the proposed procedure might provide an effective means to bridge different spatial and temporal scales in a unified multiscale modeling framework, and facilitate the application of nanoscale research results to engineering practice.
Power sources capable of supplying tens of watts are needed for a wide variety of applications including portable electronics, sensors, micro aerial vehicles, and mini-robotics systems. The utility of these devices is often limited by the energy and power density capabilities of batteries. A small combustion engine using liquid hydrocarbon fuel could potentially increase both power and energy density by an order of magnitude or more. This report describes initial development work on a meso-scale external combustion engine based on the Stirling cycle. Although other engine designs perform better at macro-scales, we believe the Stirling engine cycle is better suited to small-scale applications. The ideal Stirling cycle requires efficient heat transfer. Consequently, unlike other thermodynamic cycles, the high heat transfer rates that are inherent with miniature devices are an advantage for the Stirling cycle. Furthermore, since the Stirling engine uses external combustion, the combustor and engine can be scaled and optimized semi-independently. Continuous combustion minimizes issues with flame initiation and propagation. It also allows consideration of a variety of techniques to promote combustion that would be difficult in a miniature internal combustion engine. The project included design and fabrication of both the engine and the combustor. Two engine designs were developed. The first used a cylindrical piston design fabricated with conventional machining processes. The second design, based on the Wankel rotor geometry, was fabricated by through-mold electroforming of nickel in SU8 and LIGA micromolds. These technologies provided the requisite precision and tight tolerances needed for efficient micro-engine operation. Electroformed nickel is ideal for micro-engine applications because of its high strength and ductility. A rotary geometry was chosen because its planar geometry was more compatible with the fabrication process. SU8 lithography provided rapid prototypes to verify the design. A final high precision engine was created via LIGA. The micro-combustor was based on an excess enthalpy concept. Development of a micro-combustor included both modeling and experiments. We developed a suite of simulation tools both in support of the design of the prototype combustors, and to investigate more fundamental aspects of combustion at small scales. Issues of heat management and integration with the micro-scale Stirling engine were pursued using CFD simulations. We found that by choice of the operating conditions and channel dimensions energy conversion occurs by catalysis-dominated or catalysis-then-homogeneous phase combustion. The purpose of the experimental effort in micro-combustion was to study the feasibility and explore the design parameters of excess enthalpy combustors. The efforts were guided by the necessity for a practical device that could be implemented in a miniature power generator, or as a stand-alone device used for heat generation. Several devices were fabricated and successfully tested using methane as the fuel.
We have synthesized defect-free aluminosilicate and silicalite zeolite thin films supported on commercially available alpha and gamma alumina disk substrates. We have also built a permeation unit that can test both pure and mixed gases from room temperature to 250 C. Results indicate fluxes on the order of 10{sup -6} to 10{sup -7} mole/(m{sup 2}Pa sec) and excellent separation values for H{sub 2} or CO{sub 2}. For the Al/Si membrane: H{sub 2}/N{sub 2} {ge} 61, H{sub 2}/CO{sub 2} {ge} 80, H{sub 2}/CH{sub 4} = 7, CH{sub 4}/CO{sub 2} {ge} 11; for the TPA/Si membrane: H{sub 2}/N{sub 2} {ge} 61, H{sub 2}/CO{sub 2} {ge} 80, H{sub 2}/CH{sub 4} = 7, CH{sub 4}/CO{sub 2} {ge} 11. Our data show that we can use the adsorption ability plus the effective pore diameter of the zeolite to 'tune' the selectivity of the membrane. Another avenue of research is into bulk novel molecular sieve materials, with the goal of 'tuning' pore sizes to molecular sieving needs. A novel crystalline 12-ring microporous gallophosphate material is described.
Binding of porphyrin to murine ferrochelatase, the terminal enzyme of the heme biosynthetic pathway, is investigated by employing a set of variants harboring mutations in a putative porphyrin-binding loop. Using resonance Raman (RR) spectroscopy, the structural properties of the ferrochelatase-bound porphyrins are examined, especially with respect to the porphyrin deformation occurring in the environment of the active site. This deformation is thought to be a key step in the enzymatic insertion of ferrous iron into the porphyrin ring to make heme. Our previous RR spectroscopic studies of binding of porphyrin to murine ferrochelatase led us to propose that the wild-type enzyme induces porphyrin distortion even in the absence of the metal ion substrate. Here, we broaden this view by presenting evidence that the degree of a specific nonplanar porphyrin deformation contributes to the catalytic efficiency of ferrochelatase and its variants. The results also suggest that the conserved Trp256 (murine ferrochelatase numbering) is partially responsible for the observed porphyrin deformation. Binding of porphyrin to the ferrochelatase variants causes a decrease in the intensity of RR out-of-plane vibrational mode {gamma}{sub 15}, a saddling-like mode that is strong in the wild-type enzyme. In particular, the variant with a catalytic efficiency 1 order of magnitude lower than that of the wild-type enzyme is estimated to produce less than 30% of the wild-type saddling deformation. These results suggest that specific conserved loop residues (especially Trp256) are directly involved in the saddling of the porphyrin substrate.
In contrast to traditional terascale simulations that have known, fixed data inputs, dynamic data-driven (DDD) applications are characterized by unknown data and informed by dynamic observations. DDD simulations give rise to inverse problems of determining unknown data from sparse observations. The main difficulty is that the optimality system is a boundary value problem in 4D space-time, even though the forward simulation is an initial value problem. We construct special-purpose parallel multigrid algorithms that exploit the spectral structure of the inverse operator. Experiments on problems of localizing airborne contaminant release from sparse observations in a regional atmospheric transport model demonstrate that 17-million-parameter inversion can be effected at a cost of just 18 forward simulations with high parallel efficiency. On 1024 Alphaserver EV68 processors, the turnaround time is just 29 minutes. Moreover, inverse problems with 135 million parameters - corresponding to 139 billion total space-time unknowns - are solved in less than 5 hours on the same number of processors. These results suggest that ultra-high resolution data-driven inversion can be carried out sufficiently rapidly for simulation-based 'real-time' hazard assessment.
Accurate and efficient turbulence simulation in complex geometries is a formidable chal-lenge. Traditional methods are often limited by low accuracy and/or restrictions to simplegeometries. We explore the merger of Discontinuous Galerkin (DG) spatial discretizationswith Variational Multi-Scale (VMS) modeling, termed Local VMS (LVMS), to overcomethese limitations. DG spatial discretizations support arbitrarily high-order accuracy on un-structured grids amenable for complex geometries. Furthermore, high-order, hierarchicalrepresentation within DG provides a natural framework fora prioriscale separation crucialfor VMS implementation. We show that the combined benefits of DG and VMS within theLVMS method leads to promising new approach to LES for use in complex geometries.The efficacy of LVMS for turbulence simulation is assessed by application to fully-developed turbulent channelflow. First, a detailed spatial resolution study is undertakento record the effects of the DG discretization on turbulence statistics. Here, the localhp[?]refinement capabilites of DG are exploited to obtain reliable low-order statistics effi-ciently. Likewise, resolution guidelines for simulating wall-bounded turbulence using DGare established. We also explore the influence of enforcing Dirichlet boundary conditionsindirectly through numericalfluxes in DG which allows the solution to jump (slip) at thechannel walls. These jumps are effective in simulating the influence of the wall commen-surate with the local resolution and this feature of DG is effective in mitigating near-wallresolution requirements. In particular, we show that by locally modifying the numericalviscousflux used at the wall, we are able to regulate the near-wall slip through a penaltythat leads to improved shear-stress predictions. This work, demonstrates the potential ofthe numerical viscousflux to act as a numerically consistent wall-model and this successwarrents future research.As in any high-order numerical method some mechanism is required to control aliasingeffects due to nonlinear interactions and to ensure nonlinear stability of the method. Inthis context, we evaluate the merits of two approaches to de-aliasing -- spectralfilteringand polynomial dealiasing. While both approaches are successful, polynomial-dealiasingis found to be better suited for use in large-eddy simulation. Finally, results using LVMSare reported and show good agreement with reference direct numerical simulation therebydemonstrating the effectiveness of LVMS for wall-bounded turbulence. This success pavesthe way for future applications of LVMS to more complex turbulentflows.3
Flow-generated noise, especially rotorcraft noise has been a serious concern for bothcommercial and military applications. A particular important noise source for rotor-craft is Blade-Vortex-Interaction (BVI)noise, a high amplitude, impulsive sound thatoften dominates other rotorcraft noise sources. Usually BVI noise is caused by theunsteady flow changes around various rotor blades due to interactions with vorticespreviously shed by the blades. A promising approach for reducing the BVI noise isto use on-blade controls, such as suction/blowing, micro-flaps/jets, and smart struc-tures. Because the design and implementation of such experiments to evaluate suchsystems are very expensive, efficient computational tools coupled with optimal con-trol systems are required to explore the relevant physics and evaluate the feasibilityof using various micro-fluidic devices before committing to hardware.In this thesis the research is to formulate and implement efficient computationaltools for the development and study of optimal control and design strategies for com-plex flow and acoustic systems with emphasis on rotorcraft applications, especiallyBVI noise control problem. The main purpose of aeroacoustic computations is todetermine the sound intensity and directivity far away from the noise source. How-ever, the computational cost of using a high-fidelity flow-physics model across thefull domain is usually prohibitive and itmight also be less accurate because of thenumerical diffusion and other problems. Taking advantage of the multi-physics andmulti-scale structure of this aeroacoustic problem, we develop a multi-model, multi-domain (near-field/far-field) method based on a discontinuous Galerkin discretiza-tion. In this approach the coupling of multi-domains and multi-models is achievedby weakly enforcing continuity of normal fluxes across a coupling surface. For ourinterested aeroacoustics control problem, the adjoint equations that determine thesensitivity of the cost functional to changes in control are also solved with same ap-proach by weakly enforcing continuity ofnormal fluxes across a coupling surface.Such formulations have been validated extensively for several aeroacoustics state andcontrol problems.A multi-model based optimal control framework has been constructed and ap-plied to our interested BVI noise control problem. This model problem consists ofthe interaction of a compressible vortex with Bell AH-1 rotor blade with wall-normal3 velocity used as control on the rotor blade surface. The computational domain isdecomposed into the near-field and far-field. The near-field is obtained by numericalsolution of the Navier-Stokes equations while far away from the noise source, wherethe effect of nonlinearities is negligible, the linearized Euler equations are used tomodel the acoustic wave propagation. The BVI wave packet is targeted by definingan objective function that measures the square amplitude of pressure fluctuations inan observation region, at a time interval encompassing the dominant leading edgecompressibility waves. Our control results show that a 12dB reduction in the ob-servation region is obtained. Interestingly, the control mechanism focuses on theobservation region and only minimize the sound level in that region at the expense ofother regions. The vortex strength and trajectory get barely changed. However, theoptimal control does alter the interaction of the vortical and potential fields, whichis the source of BVI noise. While this results in a slight increase in drag, there is asignificant reduction in the temporal gradient of lift leading to a reduction in BVIsound levels.4
Catamount is designed to be a low overhead operating system for a parallel computing environment. Functionality is limited to the minimum set needed to run a scientific computation. The design choices and implementations will be presented. A massively parallel processor (MPP), high performance computing (HPC) system is particularly sensitive to operating system overhead. Traditional, multi-purpose, operating systems are designed to support a wide range of usage models and requirements. To support the range of needs, a large number of system processes are provided and are often interdependent on each other. The overhead of these processes leads to an unpredictable amount of processor time available to a parallel application. Except in the case of the most embarrassingly parallel of applications, an MPP application must share interim results with its peers before it can make further progress. These synchronization events are made at specific points in the application code. If one processor takes longer to reach that point than all the other processors, everyone must wait. The overall finish time is increased. Sandia National Laboratories began addressing this problem more than a decade ago with an architecture based on node specialization. Sets of nodes in an MPP are designated to perform specific tasks, each running an operating system best suited to the specialized function. Sandia chose to not use a multi-purpose operating system for the computational nodes and instead began developing its first light weight operating system, SUNMOS, which ran on the compute nodes on the Intel Paragon system. Based on its viability, the architecture evolved into the PUMA operating system. Intel ported PUMA to the ASCI Red TFLOPS system, thus creating the Cougar operating system. Most recently, Cougar has been ported to Cray's XT3 system and renamed to Catamount. As the references indicate, there are a number of descriptions of the predecessor operating systems. While the majority of those discussions still apply to Catamount, this paper takes a fresh look at the architecture as it is currently implemented.