Search NASA⌕ Search

SEARCH · Search NASA

Results for “shape representation”

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 559 records · Page 31

Multifidelity multiobjective optimization for wake-steering strategies

Abstract. Wake steering is an emerging wind power plant control strategy where upstream turbines are intentionally yawed out of perpendicular alignment with the incoming wind, thereby “steering” wakes away from downstream turbines. However, trade-offs between the gains in power production and fatigue loads induced by this control strategy are the subject of continuing investigation. In this study, we present a multifidelity multiobjective optimization approach for exploring the Pareto front of trade-offs between power and loading during wake steering. A large eddy simulation is used as the high-fidelity model, where an actuator line representation is used to model wind turbine blades and a rainflow-counting algorithm is used to compute damage equivalent loads. A coarser simulation with a simpler loads model is employed as a supplementary low-fidelity model. Multifidelity Bayesian optimization is performed to iteratively learn both a surrogate of the low-fidelity model and an additive discrepancy function, which maps the low-fidelity model to the high-fidelity model. Each optimization uses the expected hypervolume improvement acquisition function, weighted by the total cost of a proposed model evaluation in the multifidelity case. The multifidelity approach is able to capture the logit function shape of the Pareto frontier at a computational cost only 30 % that of the single-fidelity approach. Additionally, we provide physical insights into the vortical structures in the wake that contribute to the Pareto front shape.

17 WIND ENERGY↗

Precision calculation of above-threshold multiphoton ionization in intense short-wavelength laser fields: The momentum-space approach and time-dependent generalized pseudospectral method

We present an approach in momentum (P) space for the accurate study of multiphoton and above-threshold ionization (ATI) dynamics of atomic systems driven by intense laser fields. In this approach, the electron wave function is calculated by solving the P-space time-dependent Schroedinger equation (TDSE) in a finite P-space volume under a simple zero asymptotic boundary condition. The P-space TDSE is propagated accurately and efficiently by means of the time-dependent generalized pseudospectral method with optimal momentum grid discretization and a split-operator time propagator in the energy representation. The differential ionization probabilities are calculated directly from the continuum-state wave function obtained by projecting the total electron wave function onto the continuum-state subspace using the projection operator constructed by the continuum eigenfunctions of the unperturbed Hamiltonian. As a case study, we apply this approach to the nonperturbative study of the multiphoton and ATI dynamics of a hydrogen atom exposed to intense short-wavelength laser fields. High-resolution photoelectron energy-angular distribution and ATI spectra have been obtained. We find that with the increase of the laser intensity, the photoelectron energy-angular distribution changes from circular to dumbbell shaped and is squeezed along the laser field direction. We also explore the change of the maximum photoelectron energy with laser intensity and strong-field atomic stabilization phenomenon in detail.

74 ATOMIC AND MOLECULAR PHYSICS↗

Functional and structural mapping of human cerebral cortex: solutions are in the surfaces

The human cerebral cortex is notorious for the depth and irregularity of its convolutions and for its variability from one individual to the next. These complexities of cortical geography have been a chronic impediment to studies of functional specialization in the cortex. In this report, we discuss ways to compensate for the convolutions by using a combination of strategies whose common denominator involves explicit reconstructions of the cortical surface. Surface-based visualization involves reconstructing cortical surfaces and displaying them, along with associated experimental data, in various complementary formats (including three-dimensional native configurations, two-dimensional slices, extensively smoothed surfaces, ellipsoidal representations, and cortical flat maps). Generating these representations for the cortex of the Visible Man leads to a surface-based atlas that has important advantages over conventional stereotaxic atlases as a substrate for displaying and analyzing large amounts of experimental data. We illustrate this by showing the relationship between functionally specialized regions and topographically organized areas in human visual cortex. Surface-based warping allows data to be mapped from individual hemispheres to a surface-based atlas while respecting surface topology, improving registration of identifiable landmarks, and minimizing unwanted distortions. Surface-based warping also can aid in comparisons between species, which we illustrate by warping a macaque flat map to match the shape of a human flat map. Collectively, these approaches will allow more refined analyses of commonalities as well as individual differences in the functional organization of primate cerebral cortex.

Review↗

Modeling Deformable Linear Objects for Autonomous Robotic Outfitting of Lunar Surface Systems

This paper presents structural models of deformable linear objects (DLOs). DLOs are a subclass of deformable objects that encompasses common outfitting elements such as cables and ropes. Models are validated through hardware experiments, and integration in a robotic autonomy architecture for space environments is discussed. A persistent human presence on the lunar surface is one of the next major milestones in space exploration. This requires the development of robust extraplanetary construction technologies including structures and materials modeling and robotic systems. Previous robotic construction technology development has primarily focused on structural assembly, with significantly less focus on robotically performed outfitting tasks to instantiate subsystems providing power, data, life support, etc. These tasks involve manipulation of highly flexible elements, which are difficult to model, such as cable harnesses, ropes, and hoses. Robotic manipulation of DLOs, especially cable harnesses, is an active area of research as cable harnesses are essential for providing power and data to space assets. DLO models that can be used for robot manipulator trajectory generation are necessary for autonomous operation of lunar infrastructure. There are many proposed methods for modeling DLOs, and they primarily fall into three types: 1) discrete model-based, 2) continuum model-based, and 3) Neural Network-based. These types each have pros and cons, and the tradeoff between model accuracy and computational speed informs which type should be used. An understanding of this trade-off is imperative for real-time control of autonomous systems. High computational requirements reduce the speed of the model, making real-time control difficult, while accuracy is critical to preventing collisions. Discrete models, such as a mass-spring multibody representation, require relatively few calculations, and accuracy is directly tied to the step size of the discretization. Continuum models, such as a B-spline representation or a Cosserat rod model (a mix of continuous and discrete), are more informed of the structural properties of the cable and are much more accurate than a rigid body mass-spring model, but at significant computational cost. A Neural Network approach can provide an online solution with very few computational steps, but properly generating training data can be difficult and validation for an in-space application is not trivial. This paper explores the trade-off between different modeling approaches and compares accuracy and computational speed/complexity of the three types mentioned above. Model accuracy is evaluated using a cable in a static configuration. True cable shape is obtained using a depth camera for RGB images and point-cloud segmentation. The purpose of this experiment is to evaluate the trade-offs of different approaches to the DLO modeling problem. Understanding the tradeoffs between different cable modeling techniques paves the way for developing robotic control and planning architectures necessary for real-time manipulation of DLOs for lunar infrastructure outfitting. Real-time control is required for robotic systems to be able to actively manipulate a cable in a harsh environment where model and sensor errors compound, and environmental conditions can cause significant disturbances. Cable routing must be performed in areas with high density of objects/obstacles: through truss structures, near solar panels or mirror arrays, next to bundles of electrical equipment. Understanding the best way to plan and manipulate a cable without disrupting the environment or damaging the cable is imperative to robotic outfitting operations on the lunar surface.

Amy M Quartaro↗

Off-the-shelf real-time monitoring of satellite constellations in a visual 3-D environment

The multimission spacecraft analysis system (MSAS) data monitor is a generic software product for future real-time data monitoring and analysis. The system represents the status of a satellite constellation through the shape, color, motion and position of graphical objects floating in a three dimensional virtual reality environment. It may be used for the monitoring of large volumes of data, for viewing results in configurable displays, and for providing high level and detailed views of a constellation of monitored satellites. It is considered that the data monitor is an improvement on conventional graphic and text-based displays as it increases the amount of data that the operator can absorb in a given period, and can be installed and configured without the requirement for software development by the end user. The functionality of the system is described, including: the navigation abilities; the representation of alarms in the cybergrid; limit violation; real-time trend analysis, and alarm status indication.

Schwuttke, Ursula M.↗

Structure and Parameter Uncertainty in Centennial Projections of Forest Community Structure and Carbon Cycling

Secondary forest regrowth shapes community succession and biogeochemistry for decades, including in the Upper Great Lakes region. Vegetation models encapsulate our understanding of forest function, and whether models can reproduce multi‐decadal succession patterns is an indication of our ability to predict forest responses to future change. We test the ability of a vegetation model to simulate C cycling and community composition during 100 years of forest regrowth following stand‐replacing disturbance, asking (a) Which processes and parameters are most important to accurately model Upper Midwest forest succession? (b) What is the relative importance of model structure versus parameter values to these predictions? We ran ensembles of the Ecosystem Demography model v2.2 with different representations of processes important to competition for light. We compared the magnitude of structural and parameter uncertainty and assessed which sub‐model–parameter combinations best reproduced observed C fluxes and community composition. On average, our simulations underestimated observed net primary productivity (NPP) and leaf area index (LAI) after 100 years and predicted complete dominance by a single plant functional type (PFT). Out of 4,000 simulations, only nine fell within the observed range of both NPP and LAI, but these predicted unrealistically complete dominance by either early hardwood or pine PFTs. A different set of seven simulations were ecologically plausible but under‐predicted observed NPP and LAI. Parameter uncertainty was large; NPP and LAI ranged from ~0% to >200% of their mean value, and any PFT could become dominant. The two parameters that contributed most to uncertainty in predicted NPP were plant–soil water conductance and growth respiration, both unobservable empirical coefficients. We conclude that (a) parameter uncertainty is more important than structural uncertainty, at least for ED‐2.2 in Upper Midwest forests and (b) simulating both productivity and plant community composition accurately without physically unrealistic parameters remains challenging for demographic vegetation models.

canopy radiative transfer↗

Theoretical characterization of the potential energy surface for H + N2 yields HN2. II - Computed points to define a global potential

A previous calculation for H + N2 (Walch et al., 1989) focused on the minimum energy path (MEP) region of the potential energy surface and on estimates of the lifetime of the HN2 species. In this paper, energies computed at geometries selected to permit a global representation of the potential energy surface (PES) are reported. As in the previous work, the calculations were performed using the complete active space self-consistent field/externally contracted configuration interaction method. The surface was characterized using the same basis set as in the previous paper except that an improved contraction of the H s-basis is used. Calculations with a larger basis set were carried out along an approximate MEP obtained with the smaller basis set. The new PES exhibits a sharp curvature, which was not present in the previous calculations, and has a slightly narrower and smaller barrier to dissociation. Saddle points for H atom exchange via collinear and T shaped HN2 complexes are also reported.

Walch, Stephen P.↗

Validation and Simulation of ARES I Scale Model Acoustic Test -1- Pathfinder Development

The Ares I Scale Model Acoustics Test (ASMAT) is a series of live-fire tests of scaled rocket motors meant to simulate the conditions of the Ares I launch configuration. These tests have provided a well documented set of high fidelity measurements useful for validation including data taken over a range of test conditions and containing phenomena like Ignition Over-Pressure and water suppression of acoustics. To take advantage of this data, a digital representation of the ASMAT test setup has been constructed and test firings of the motor have been simulated using the Loci/CHEM computational fluid dynamics software. Within this first of a series of papers, results from ASMAT simulations with the rocket in a held down configuration and without water suppression have then been compared to acoustic data collected from similar live-fire tests to assess the accuracy of the simulations. Detailed evaluations of the mesh features, mesh length scales relative to acoustic signals, Courant-Friedrichs-Lewy numbers, and spatial residual sources have been performed to support this assessment. Results of acoustic comparisons have shown good correlation with the amplitude and temporal shape of pressure features and reasonable spectral accuracy up to approximately 1000 Hz. Major plume and acoustic features have been well captured including the plume shock structure, the igniter pulse transient, and the ignition overpressure. Finally, acoustic propagation patterns illustrated a previously unconsidered issue of tower placement inline with the high intensity overpressure propagation path.

Putnam, G. C.↗

The Mathematical Foundations of the Science of Cities

In this chapter, we describe graph-theoretic representations of infrastructure and social processes in urban environments and trace the development of these fields from the perspective of mathematics, social science, and urban planning. We follow the historical development of two different perspectives on cities and urban planning – one in which infrastructure and urban form is the primary focus and another which made people and social processes the primary focus. These perspectives can be traced through their application of concurrently developing mathematical techniques in graph theory, network science, and social network analysis. These different perspectives are now becoming integrated into a more mathematically grounded understanding of cities as coupled social and physical systems, in which the social life of a city shapes and defines the infrastructure that is built and a city’s infrastructure and physical form also shape the lives and communities of its residents.

Brelsford, Christa↗

Advanced Finite-Volume Numerics and Source Term Assumptions for Kernel and G-Equation Modelling of Propane/Air Flames

Here G-Equation models represent propagating flame fronts with an implicit two-dimensional surface representation (level-set). Level-set methods are fast, as transport source terms for the implicit surface can be solved with finite-volume operators on the finite-volume domain, without having to build the actual surface. However, they include approximations whose practical effects are not properly understood. In this study, we improved the numerics of the FRESCO CFD code’s G-Equation solver and developed a new method to simulate kernel growth using signed distance functions and the analytical sphere-mesh overlap. We analyzed their role for simulating propane/air flames, using three well-established constant-volume configurations: a one-dimensional, freely propagating laminar flame; a disc-shaped, constant-volume swirl combustor; and torch-jet flame development through an orifice from a two-chamber device. We tested the explicit (sub-cycled) vs. implicit formulation for the standard transport operators (advection, diffusion, compressibility). In addition to the accurate flame swept-volume method for chemistry and species source term, we developed a more accurate estimator for the burnt/unburnt split cell composition. Then, we developed a signed-distance-function (SDF) based method which provides a more stable reinitialization of the level-set field at every time-step. We found that simplifying assumptions common to several G-Equation implementations, for straightforward terms such as compressibility and advection, lead to large errors in predicting the propagation of even laminar flames, with deviations up to ~300% in simulated vs. formulated flame speed. Conversely, the enhanced numerics enabled through the SDF field reinitialization and improved chemistry source term improve simulation stability and smooth flame propagation even with significantly larger solver time-steps.

42 ENGINEERING↗

Attributes of GRB Pulses: Analysis of BATSE TTE Data

Our extensive program of modeling GRB profiles is aimed at elucidating the physical processes responsible for the burst itself, as well as possible extrinsic phenomena (e.g. time dilation) as discussed in another paper in this Symposium (Norris et al., "GRB PROFILES AS COSMIC PROBES"). We have developed special methods to extract the wealth of short time-scale information contained in the BATSE time-tag event (TTE) data. Our algorithm yields a piecewise-constant representation of the light curve -- using only the raw photon arrival times, and based on Bayesian change-point methods. This representation in effect lets the data determine the bin size and location, and avoids unwanted effects due to arbitrary choices of the bin parameters. We have determined widths, separations, and amplitudes of pulses contained in the bursts, without invoking a specific pulse model. The effect of cosmic time dilation can be easily seen in a direct plot of amplitude vs. time scale for individual pulses, without the need to lump the data into a small number of brightness classes. We are also performing noise equalization on these data (to reduce a well-known bias of pulse width as a function of signal-to-noise ratio), as well as fits of parametric pulse-shape models -- including explicit energy dependence of the pulse parameters. Such refinements are expected to improve the quality and physical significance of these results.

Norris, J. P.↗

Integrating nonlinear aerodynamic and structural analysis for a complete fighter configuration

The coupling of a nonlinear aerodynamics program with a structural analysis program to include the effects of static aeroelasticity in early preliminary design studies is described. A nonlinear, full potential aerodynamics method with capability to model geometric details of a complete aircraft in supersonic flow is used. The deflections of the lifting surfaces are calculated using an equivalent plate structural representation which can readily accommodate the changes in stiffness and geometric properties required during the preliminary design process. An iterative solution procedure is used to obtain consistent aerodynamic loads and structural deflections at the specified flight conditions. The volume of data transmitted between programs is minimized. The procedure is applied to a complete aircraft and the numerical results illustrate the aeroelastic effects on pressure distribution as well as total forces and moments. During this design study, the thickness distribution of the wing cover skins was initially sized based on rigid loads and subsequently resized under aeroelastic loads. Comparisons are made between these nonlinear aeroelastic results and results obtained from linear aerodynamic methods applied to a rigid shape during conceptual design studies.

Tatum, Kenneth E.↗

Using Clustering to Establish Climate Regimes from PCM Output

A multivariate statistical clustering technique--based on the k-means algorithm of Hartigan has been used to extract patterns of climatological significance from 200 years of general circulation model (GCM) output. Originally developed and implemented on a Beowulf-style parallel computer constructed by Hoffman and Hargrove from surplus commodity desktop PCs, the high performance parallel clustering algorithm was previously applied to the derivation of ecoregions from map stacks of 9 and 25 geophysical conditions or variables for the conterminous U.S. at a resolution of 1 sq km. Now applied both across space and through time, the clustering technique yields temporally-varying climate regimes predicted by transient runs of the Parallel Climate Model (PCM). Using a business-as-usual (BAU) scenario and clustering four fields of significance to the global water cycle (surface temperature, precipitation, soil moisture, and snow depth) from 1871 through 2098, the authors' analysis shows an increase in spatial area occupied by the cluster or climate regime which typifies desert regions (i.e., an increase in desertification) and a decrease in the spatial area occupied by the climate regime typifying winter-time high latitude perma-frost regions. The patterns of cluster changes have been analyzed to understand the predicted variability in the water cycle on global and continental scales. In addition, representative climate regimes were determined by taking three 10-year averages of the fields 100 years apart for northern hemisphere winter (December, January, and February) and summer (June, July, and August). The result is global maps of typical seasonal climate regimes for 100 years in the past, for the present, and for 100 years into the future. Using three-dimensional data or phase space representations of these climate regimes (i.e., the cluster centroids), the authors demonstrate the portion of this phase space occupied by the land surface at all points in space and time. Any single spot on the globe will exist in one of these climate regimes at any single point in time. By incrementing time, that same spot will trace out a trajectory or orbit between and among these climate regimes (or atmospheric states) in phase (or state) space. When a geographic region enters a state it never previously visited, a climatic change is said to have occurred. Tracing out the entire trajectory of a single spot on the globe yields a 'manifold' in state space representing the shape of its predicted climate occupancy. This sort of analysis enables a researcher to more easily grasp the multivariate behavior of the climate system.

Oglesby, Robert↗

Guided Wave Delamination Detection and Quantification With Wavefield Data Analysis

Unexpected damage can occur in aerospace composites due to impact events or material stress during off-nominal loading events. In particular, laminated composites are susceptible to delamination damage due to weak transverse tensile and inter-laminar shear strengths. Developments of reliable and quantitative techniques to detect delamination damage in laminated composites are imperative for safe and functional optimally-designed next-generation composite structures. In this paper, we investigate guided wave interactions with delamination damage and develop quantification algorithms by using wavefield data analysis. The trapped guided waves in the delamination region are observed from the wavefield data and further quantitatively interpreted by using different wavenumber analysis methods. The frequency-wavenumber representation of the wavefield shows that new wavenumbers are present and correlate to trapped waves in the damage region. These new wavenumbers are used to detect and quantify the delamination damage through the wavenumber analysis, which can show how the wavenumber changes as a function of wave propagation distance. The location and spatial duration of the new wavenumbers can be identified, providing a useful means not only for detecting the presence of delamination damage but also allowing for estimation of the delamination size. Our method has been applied to detect and quantify real delamination damage with complex geometry (grown using a quasi-static indentation technique). The detection and quantification results show the location, size, and shape of the delamination damage.

Tian, Zhenhua↗

Horizontal Variability of Water and Its Relationship to Cloud Fraction near the Tropical Tropopause: Using Aircraft Observations of Water Vapor to Improve the Representation of Grid-scale Cloud Formation in GEOS-5

Large-scale models such as GEOS-5 typically calculate grid-scale fractional cloudiness through a PDF parameterization of the sub-gridscale distribution of specific humidity. The GEOS-5 moisture routine uses a simple rectangular PDF varying in height that follows a tanh profile. While below 10 km this profile is informed by moisture information from the AIRS instrument, there is relatively little empirical basis for the profile above that level. ATTREX provides an opportunity to refine the profile using estimates of the horizontal variability of measurements of water vapor, total water and ice particles from the Global Hawk aircraft at or near the tropopause. These measurements will be compared with estimates of large-scale cloud fraction from CALIPSO and lidar retrievals from the CPL on the aircraft. We will use the variability measurements to perform studies of the sensitivity of the GEOS-5 cloud-fraction to various modifications to the PDF shape and to its vertical profile.

climate modeling↗

Acoustic-based monitoring and machine learning of component status for microreactor applications

This report provides a description and assessment of recent efforts to couple acoustic-based experimental measurements and characterization with machine learning models in order to enhance structural health monitoring capabilities for nuclear microreactors. With resilient embedded sensors in development by others supported by programs funded by the US Department of Energy’s Office of Nuclear Energy, the work described herein builds upon ongoing efforts to improve non-destructive testing technology that relates measured acoustic signatures to component stresses and/or structural defects, using a combination of new experimental measurements and machine learning architectures. The experimental procedure remained similar to that developed for the previous year’s demonstration of damage detection by the authors, with the same damaged sample tested under similar applied stress conditions. Notably, a new mounting fixture was designed and implemented to improve measurement consistency and a more sophisticated laser Doppler vibrometer was employed to make high-fidelity vibration measurements. Two nominally identical sets of training data were collected for each experimental setup to better understand the repeatability of the experiment and to better test the generality of trained neural network models. Additionally, we obtained new high-quality 3D mode shapes of the damaged test article at various stress and excitation levels, providing greater insights into the physical response of the sample during testing. Previously, we demonstrated that a machine learning model based on a convolutional neural network can predict structural details of an artificially introduced interface (intact, rough cut, smooth cut), and the applied torque level. In this study, we have transitioned to graph-based neural network architectures to better develop and test a flexible framework that is more suitable to being transferred away from controlled benchtop experiments and into more applied settings where less-structured data inputs may be expected. In general, performance testing of a graph neural network on frequency-domain representations of the data indicates strong and consistent identification of test conditions for datasets recorded on damaged components. With goals of predicting damage location and other changing experimental conditions using limited datasets, predictive models using a graph neural network architecture correctly predicted the applied torque level with an accuracy of 85% using only a single measurement point and predicted within one torque level in 95% of test windows. Predictions of damage location had limited success due to the symmetry and minimal number of the damage scenarios presented during model training. Results were ambiguous as to whether the model could detect the location of the artificial damage, or if it was instead learning the location of a given measurement point on the part and subsequently detecting which points were closest to the location of the damage. This finding will be factored into upcoming planned work on damaged graphite components, where new experimental tests with a larger number and variety of damage scenarios are expected to provide improved validation of recent developments in monitoring methodology.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Knowledge Oriented Graph Unified Transformer (KOGUT) v0.1

KOGUT — Knowledge Oriented Graph Unified Transformer KOGUT implements the Relational Graph Transformer (RelGT) architecture for knowledge graph link prediction in biological domains, with a primary focus on microbial growth media prediction. While the original RelGT (arXiv:2505.10960) targets relational tables, time series, and multi-table databases, KOGUT adapts this architecture for heterogeneous biological knowledge graphs, providing first-in-class AI predictive models for microbial cultivation. Key Adaptations Beyond Original RelGT: - Knowledge Graph Focus: Applied to biological KGs with semantic node types (taxa, chemicals, media, phenotypes, environments) versus generic relational database tables, trained on the KG-Microbe knowledge graph (1.3M entities, 2.9M edges, 24 relation types). - Multimodal Node Encoding: Integrates node labels, categories, descriptions, and synonyms from KG metadata through learned embedding layers—adapting relational column features to graph node attributes with textual semantics. - Extended K-Hop Subgraph Strategy: Optimized neighborhood sampling (3-hop default, configurable up to 200 nodes) tuned for sparse biological networks, building on the original local-global attention framework with biological relation preservation. - Biolink Predicate Preservation: Type-specific transformations for 24 biological edge semantics (occurs_in, consumes, produces, has_phenotype, subclass_of) beyond standard relational foreign keys, enabling multi-relation link prediction. - Inductive Learning Support: Enables zero-shot predictions for novel taxa through feature-based embeddings (temperature, oxygen requirements, gram stain, cell shape), extending the original transductive relational benchmark scope to uncultured microorganisms. CheapSOTA Performance Optimizations (This Distribution): - VQ-EMA Centroid Attention: Vector quantization with exponential moving average for improved global context modeling (+5-10% MRR improvement). - HDF5 Precomputed Data Loading: One-time preprocessing of k-hop subgraphs to eliminate redundant graph traversals (2-5× training speedup). - Distributed Data Parallel Training: Multi-GPU support for scaling to larger knowledge graphs (tested on 4× NVIDIA A100 GPUs at NERSC Perlmutter). - Mixed Precision Training: Automatic mixed precision (AMP) for memory efficiency and faster training. Advantages Over Standard Knowledge Graph Embedding Models: Combines RelGT's proven multi-element tokenization (features, type, hop, structure) with graph-native biological representations, enabling interpretable link prediction across heterogeneous entities that standard embedding models (TransE, RotatE, ComplEx) and table-based transformers cannot directly model. Achieves near-perfect performance on microbial growth media prediction (MRR: 0.9966, Precision@1: 0.9932, Hit@10: 1.0000) while maintaining explainability through attention-based reasoning over biological pathways. Training Data: - KG-Microbe merged knowledge graph: 1,379,337 nodes, 2,960,472 edges - 24 biological relation types including taxonomic hierarchies, metabolic interactions, phenotype associations, and environmental relationships - Primary prediction task: Growth media suitability for microbial taxa (biolink:occurs_in, 50K edges) - Multi-relation capability: Predicts links for any of the 24 relation types, including chemical consumption/production, phenotype associations, and taxonomic classification Citation: Original RelGT Architecture: Dwivedi et al., "Relational Graph Transformer", arXiv:2505.10960, 2025 KOGUT Implementation: Knowledge Oriented Graph Unified Transformer for Microbial Growth Media Prediction Developed at Lawrence Berkeley National Laboratory (LBNL) Trained on NERSC Perlmutter supercomputer

Joachimiak, Marcin [Lawrence Berkeley National Lab↗

Shaping Diffraction-Grating Grooves to Optimize Efficiency

A method of shaping diffraction-grating grooves to optimize the spectral efficiency, spectral range, and image quality of a spectral imaging instrument is under development. The method is based on the use of an advanced design algorithm to determine the possibly complex shape of grooves needed to obtain a desired efficiency-versus-wavelength response (see figure). Then electron- beam fabrication techniques are used to realize the required groove shape. The method could be used, for example, to make the spectral efficiency of the grating in a given wavelength range proportional to the inverse of the spectral efficiency of a photodetector array so that the overall spectral efficiency of the combination of the grating and the photodetector array would be flat. The method has thus far been applied to one-dimensional gratings only, but in principle, it is also applicable to two-dimensional gratings. The algorithm involves calculations in the spatial-frequency domain. The spatial-frequency spectrum of a grating is represented as a diffraction-order spectral-peak-width function multiplied by an efficiency function for a single grating groove. This representation affords computational efficiency and accuracy by making it possible to consider only the response from one grating groove (one period of the grating), instead of from the whole grating area, in determining the response from the entire grating. This combination of efficiency and accuracy is crucial for future extensions of the algorithm to two-dimensional designs and to designs in which polarization must also be taken into account. The algorithm begins with the definition of target values of relative efficiency that represent the desired spectral response of the grating in certain spectral frequencies calculated from the diffraction order and wavelength. The grating period is divided into a number of cells - typically, 100. The phase contribution from each cell is determined from the phase of the incident electromagnetic wave and the height of the grating surface in the cell. The total contribution from all cells to each target value is then calculated. Then a method known to specialists as the optimum-rotation-angle method is used to adjust the height of each cell so that the total response from all cells is optimized. The computation is iterative and continues until the desired response is obtained. In the event that the desired response is unphysical, the algorithm nevertheless strives to generate a grating-grove profile for which the response approximates the desired one as closely as possible.

Backlund, John↗