Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “stochastic models”

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 1,747 records · Page 97

Probabilistic Residual Strength Model Developed for Life Prediction of Ceramic Matrix Composites

For the next generation of reusable launch vehicles, NASA is investigating introducing ceramic matrix composites (CMCs) in place of current superalloys for structural propulsion applications (e.g., nozzles, vanes, combustors, and heat exchangers). The higher use temperatures of CMCs will reduce vehicle weight by eliminating and/or reducing cooling system requirements. The increased strength-to-weight ratio of CMCs relative to superalloys further enhances their weight savings potential. However, in order to provide safe designs for components made of these new materials, a comprehensive life prediction methodology for CMC structures needs to be developed. A robust methodology for lifing composite structures has yet to be adopted by the engineering community. Current industry design practice continues to utilize deterministic empirically based models borrowed from metals design for predicting material life capabilities. The deterministic nature of these models inadequately addresses the stochastic character of brittle composites, and their empirical reliance makes predictions beyond the experimental test conditions a risky extrapolation. A team of engineers at the NASA Glenn Research Center has been developing a new life prediction engineering model. The Probabilistic Residual Strength (PRS) model uses the residual strength of the composite as its damage metric. Expected life and material strength are both considered probabilistically to account for the observed stochastic material response. Extensive experimental testing has been carried out on C/SiC (a candidate aerospace CMC material system) in a controlled 1000 ppm O2/argon environment at elevated temperatures of 800 and 1200 C. The test matrix was established to allow observation of the material behavior, characterization of the model, and validation of the model's predictive capabilities. Sample results of the validation study are illustrated in the graphs.

Thomas, David J.↗

Classical-quantum scattering

We analyze the framework recently proposed by Oppenheim et al (2023 Nat. Commun. 14; 2023 Phys. Rev. X 13 041040; arXiv:2302.07283 [gr-qc]; 2023 J. High Energy Phys. JHEP08(2023)163) to model relativistic quantum fields coupled to relativistic, classical, stochastic fields (in particular, as a model of quantum matter coupled to ‘classical gravity’). Perhaps surprisingly, we find that we can define and calculate scattering probabilities which are Lorentz-covariant and conserve total probability, at least at tree level. As a concrete example, we analyze 2→2 scattering of quantum matter mediated by a classical Yukawa field. Mapping this to a gravitational coupling in the non-relativistic limit, and assuming that we can treat large objects as point masses, we find that the simplest possible ‘classical-quantum’ gravity theory constructed this way gives predictions for 2→2 gravitational scattering which are inconsistent with simple observations of, e.g. spacecraft undergoing slingshot maneuvers. We comment on lessons learned for attempts to couple quantum matter to ‘non-quantum’ gravity, or more generally, for attempts to couple relativistic quantum and classical systems.

Carney, Daniel↗

On the Representation of Hyporheic Exchange in Models for Reactive Transport in Stream and River Corridors

Efforts to include more detailed representations of biogeochemical processes in basin-scale water quality simulation tools face the challenge of how to tractably represent mass exchange between the flowing channels of streams and rivers and biogeochemical hotspots in the hyporheic zones. Multiscale models that use relatively coarse representations of the channel network with subgrid models for mass exchange and reactions in the hyporheic zone have started to emerge to address that challenge. Two such multiscale models are considered here, one based on a stochastic Lagrangian travel time representation of advective pumping and one on multirate diffusive exchange. The two models are formally equivalent to well-established integrodifferential representations for transport of non-reacting tracers in steady stream flow, which have been very successful in reproducing stream tracer tests. Despite that equivalence, the two models are based on very different model structures and produce significantly different results in reactive transport. In a simple denitrification example, denitrification is two to three times greater for the advection-based model because the multirate diffusive model has direct connections between the stream channel and transient storage zones and an assumption of mixing in the transient storage zones that prevent oxygen levels from dropping to the point where denitrification can progress uninhibited. By contrast, the advection-based model produces distinct redox zonation, allowing for denitrification to proceed uninhibited on part of the hyporheic flowpaths. These results demonstrate that conservative tracer tests alone are inadequate for constraining representation of mass transfer in models for reactive transport in streams and rivers.

54 ENVIRONMENTAL SCIENCES↗

Highly-scalable, physics-informed GANs for learning solutions of stochastic PDEs

Uncertainty quantification for forward and inverse problems is a central challenge across physical and biomedical disciplines. We address this challenge for the problem of modeling subsurface flow at the Hanford Site by combining stochastic computational models with observational data using physics-informed GAN models. The geographic extent, spatial heterogeneity, and multiple correlation length scales of the Hanford Site require training a computationally intensive GAN model to thousands of dimensions. We develop a highly optimized implementation that scales to {\herogpucount} NVIDIA Volta GPUs. We develop a hierarchical scheme based on a multi-player game-theoretic approach for exploiting domain parallelism, map discriminators and generators to multiple GPUs, and employ efficient communication schemes to ensure training stability and convergence. Our implementation scales to {\heronodecount} nodes on the Summit supercomputer with a \weakscalingeff\% scaling efficiency, achieving peak and sustained half-precision rates of {\heropeakperf} PF/s and {\herosustainedperf} PF/s.

physics informed ML, machine learning, generative ↗

A Coupling Methodology for Mesoscale-informed Nuclear Fuel Performance Codes

This study proposes an approach for capturing the effect of microstructural evolution on reactor fuel performance by coupling a mesoscale irradiated microstructure model with a finite element fuel performance code. To achieve this, the macroscale system is solved in a parallel, fully coupled, fully-implicit manner using the preconditioned Jacobian-free Newton Krylov (JFNK) method. Within the JFNK solution algorithm, microstructure-influenced material parameters are calculated by the mesoscale model and passed back to the macroscale calculation. Due to the stochastic nature of the mesoscale model, a dynamic fitting technique is implemented to smooth roughness in the calculated material parameters. The proposed methodology is demonstrated on a simple model of a reactor fuel pellet. In the model, INL’s BISON fuel performance code calculates the steady-state temperature profile in a fuel pellet and the microstructure-influenced thermal conductivity is determined with a phase field model of irradiated microstructures. This simple multiscale model demonstrates good nonlinear convergence and near ideal parallel scalability. By capturing the formation of large mesoscale voids in the pellet interior, the multiscale model predicted the irradiation-induced reduction in the thermal conductivity commonly observed in reactors.

97 MATHEMATICS AND COMPUTING↗

Noisy homoclinic pulse dynamics

The effect of stochastic perturbations on nearly homoclinic pulse trains is considered for three model systems: a Duffing oscillator, the Lorenz-like Shimizu–Morioka model, and a co-dimension-three normal form. Using the Duffing model as an example, it is demonstrated that the main effect of noise does not originate from the neighbourhood of the fixed point, as is commonly assumed, but due to the perturbation of the trajectory outside that region. Singular perturbation theory is used to quantify this noise effect and is applied to construct maps of pulse spacing for the Shimizu–Morioka and normal form models. The dynamics of these stochastic maps is then explored to examine how noise influences the sequence of bifurcations that take place adjacent to homoclinic connections in Lorenz-like and Shilnikov-type flows.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Cosmology with many light scalar fields: Stochastic inflation and loop corrections

We explore the consequences of the existence of a very large number of light scalar degrees of freedom in the early universe. We distinguish between participator and spectator fields. The former have a small mass, and can contribute to the inflationary dynamics; the latter are either strictly massless or have a negligible VEV. In N-flation and generic assisted inflation scenarios, inflation is a cooperative phenomenon driven by N participator fields, none of which could drive inflation on its own. We review upper bounds on N, as a function of the inflationary Hubble scale H. We then consider stochastic and eternal inflation in models with N participator fields showing that individual fields may evolve stochastically while the whole ensemble behaves deterministically, and that a wide range of eternal inflationary scenarios are possible in this regime. We then compute one-loop quantum corrections to the inflationary power spectrum. These are largest with N spectator fields and a single participator field, and the resulting bound on N is always weaker than those obtained in other ways. We find that loop corrections to the N-flation power spectrum do not scale with N, and thus place no upper bound on the number of participator fields. This result also implies that, at least to leading order, the theory behaves like a composite single scalar field. In order to perform this calculation, we address a number of issues associated with loop calculations in the Schwinger-Keldysh ''in-in'' formalism.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Attitude/attitude-rate estimation from GPS differential phase measurements using integrated-rate parameters

A sequential filtering algorithm is presented for attitude and attitude-rate estimation from Global Positioning System (GPS) differential carrier phase measurements. A third-order, minimal-parameter method for solving the attitude matrix kinematic equation is used to parameterize the filter's state, which renders the resulting estimator computationally efficient. Borrowing from tracking theory concepts, the angular acceleration is modeled as an exponentially autocorrelated stochastic process, thus avoiding the use of the uncertain spacecraft dynamic model. The new formulation facilitates the use of aiding vector observations in a unified filtering algorithm, which can enhance the method's robustness and accuracy. Numerical examples are used to demonstrate the performance of the method.

Oshman, Yaakov↗

Vector Observation-Aided/Attitude-Rate Estimation Using Global Positioning System Signals

A sequential filtering algorithm is presented for attitude and attitude-rate estimation from Global Positioning System (GPS) differential carrier phase measurements. A third-order, minimal-parameter method for solving the attitude matrix kinematic equation is used to parameterize the filter's state, which renders the resulting estimator computationally efficient. Borrowing from tracking theory concepts, the angular acceleration is modeled as an exponentially autocorrelated stochastic process, thus avoiding the use of the uncertain spacecraft dynamic model. The new formulation facilitates the use of aiding vector observations in a unified filtering algorithm, which can enhance the method's robustness and accuracy. Numerical examples are used to demonstrate the performance of the method.

Oshman, Yaakov↗

Quantification of errors in large-eddy simulations of a spatially evolving mixing layer using polynomial chaos

A stochastic approach based on generalized polynomial chaos (gPC) is used to quantify the error in large-eddy simulation (LES) of a spatially evolving mixing layer flow and its sensitivity to different simulation parameters, viz., the grid stretching in the streamwise and lateral directions and the subgrid-scale (SGS) Smagorinsky model constant (C{sub S}). The error is evaluated with respect to the results of a highly resolved LES and for different quantities of interest, namely, the mean streamwise velocity, the momentum thickness, and the shear stress. A typical feature of the considered spatially evolving flow is the progressive transition from a laminar regime, highly dependent on the inlet conditions, to a fully developed turbulent one. Therefore, the computational domain is divided in two different zones (inlet dependent and fully turbulent) and the gPC error analysis is carried out for these two zones separately. An optimization of the parameters is also carried out for both these zones. For all the considered quantities, the results point out that the error is mainly governed by the value of the C{sub S} constant. At the end of the inlet-dependent zone, a strong coupling between the normal stretching ratio and the C{sub S} value is observed. The error sensitivity to the parameter values is significantly larger in the inlet-dependent upstream region; however, low-error values can be obtained in this region for all the considered physical quantities by an ad hoc tuning of the parameters. Conversely, in the turbulent regime the error is globally lower and less sensitive to the parameter variations, but it is more difficult to find a set of parameter values leading to optimal results for all the analyzed physical quantities. A similar analysis is also carried out for the dynamic Smagorinsky model, by varying the grid stretching ratios. Comparing the databases generated with the different subgrid-scale models, it is possible to observe that the error cost function computed for the streamwise velocity and for the momentum thickness is not significantly sensitive to the used SGS closure. Conversely, the prediction of the shear stress is much more accurate when using a dynamic subgrid-scale model and the variance of the error is lower in magnitude.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Minimalist Model of Ice Microphysics in Mixed-phase Stratiform Clouds

The question of whether persistent ice crystal precipitation from super cooled layer clouds can be explained by time-dependent, stochastic ice nucleation is explored using an approximate, analytical model, and a large-eddy simulation (LES) cloud model. The updraft velocity in the cloud defines an accumulation zone, where small ice particles cannot fall out until they are large enough, which will increase the residence time of ice particles in the cloud. Ice particles reach a quasi-steady state between growth by vapor deposition and fall speed at cloud base. The analytical model predicts that ice water content (wi) has a 2.5 power law relationship with ice number concentration ni. wi and ni from a LES cloud model with stochastic ice nucleation also confirm the 2.5 power law relationship. The prefactor of the power law is proportional to the ice nucleation rate, and therefore provides a quantitative link to observations of ice microphysical properties.

clouds↗

Modeling, Real-Time Estimation, and Identification of UWB Indoor Wireless Channels

Stochastic differential equations (SDEs) are used to model ultrawideband (UWB) indoor wireless channels. We show that the impulse responses for time-varying indoor wireless channels can be approximated in a mean-square sense as close as desired by impulse responses that can be realized by SDEs. The state variables represent the inphase and quadrature components of the UWB channel. The expected maximization and extended Kalman filter are employed to recursively identify and estimate the channel parameters and states, respectively, from online received signal strength measured data. Both resolvable and nonresolvable multipath received signals are considered and represented as small-scaled Nakagami fading. The proposed models together with the estimation algorithm are tested using UWB indoor measurement data demonstrating the method’s viability and the results are presented.

Olama, Mohammed M.↗

Modeling Streamflow and Sediment Transport Under Current and Future Climates at the West Valley Site - 20231

Performance Assessment models often require inputs representing specific hydrologic processes such as streamflow and sediment flux. This paper details the development of a process-level hydrologic model and its use in a Probabilistic Performance Assessment (PPA) system-level model for the West Valley Site in western New York. For this PPA model, a watershed model was developed using SWAT (Soil and Water Assessment Tool) and used to evaluate water and sediment fluxes in nearby streams draining the Site. A particular challenge of probabilistic modeling over long periods of time is scaling parameters appropriately so that outcomes are not grossly over- or underestimated. This challenge is especially important when the spatial and temporal scales of process and system level models differ. In this instance, the SWAT watershed model and GoldSim PPA model have similar spatial scales of stream reaches and sub-catchments within a single watershed (approximately hundreds to tens of thousands of square meters, but the temporal scales differ considerably. The SWAT model calculates streamflow and sediment transport rates at the daily scale, while the PPA model requires long-term average values that are applied over hundreds to thousands of year-long periods. As such, distributions of daily climatic parameters such as average precipitation rate, temperature, wind speed, and relative humidity were developed to force a transient SWAT model that captures the dynamic behavior of the watershed in response to daily changes in the weather. Output from the model, which includes stream flow (m{sup 3}/y) and sediment transport rate (Mg/y) is then averaged over a 100-year period as an approximation for the long-term average values which can be applied in GoldSim. A single SWAT model was run 5000 times with varying climatic inputs and deterministic soil, land use, and elevation data, and the range of outputs were compiled into a distribution which can be implemented stochastically into the PPA Model. Additionally, this paper includes a high-level discussion of how changes in future climate affect weather input distributions, and a summary of how those changes in weather affect the outcomes from the SWAT model and the inputs into the PPA model. In this analysis, future climate distributions were developed based on climate research documenting the expected changes to precipitation and temperature in western New York. The 5,000 SWAT simulations were then repeated with these future climate distributions, and the resulting output was implemented into the PPA Model as hydrologic conditions under a future climate. In the current version of the PPA Model, hydrologic parameters change linearly from the 'current climate' conditions to the 'future climate' conditions over a period of 100 years, after which they remain constant. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Hybrid stochastic synapses enabled by scaled ferroelectric field-effect transistors

Achieving brain-like density and performance in neuromorphic computers necessitates scaling down the size of nanodevices emulating neuro-synaptic functionalities. However, scaling nanodevices results in reduction of programming resolution and emergence of stochastic non-idealities. While prior work has mainly focused on binary transitions, in this work, we leverage the stochastic switching of a three-state ferroelectric field-effect transistor to implement a long-term and short-term two-tier stochastic synaptic memory with a single device. Experimental measurements are performed on a scaled 28 nm high-k metal gate technology-based device to develop a probabilistic model of the hybrid stochastic synapse. In addition to the advantage of ultra-low programming energies afforded by scaling, our hardware–algorithm co-design analysis reveals the efficacy of the two-tier memory in comparison to binary stochastic synapses in on-chip learning tasks—paving the way for algorithms exploiting multi-state devices with probabilistic transitions beyond deterministic ones.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Dependability and performability analysis

Several practical issues regarding specifications and solution of dependability and performability models are discussed. Model types with and without rewards are compared. Continuous-time Markov chains (CTMC's) are compared with (continuous-time) Markov reward models (MRM's) and generalized stochastic Petri nets (GSPN's) are compared with stochastic reward nets (SRN's). It is shown that reward-based models could lead to more concise model specifications and solution of a variety of new measures. With respect to the solution of dependability and performability models, three practical issues were identified: largeness, stiffness, and non-exponentiality, and a variety of approaches are discussed to deal with them, including some of the latest research efforts.

Trivedi, Kishor S.↗

A stepped leader model for lightning including charge distribution in branched channels

The stepped leader process in negative cloud-to-ground lightning plays a vital role in lightning protection analysis. As lightning discharge usually presents significant branched or tortuous channels, the charge distribution along the branched channels and the stochastic feature of stepped leader propagation were investigated in this paper. The charge density along the leader channel and the charge in the leader tip for each lightning branch were approximated by introducing branch correlation coefficients. In combination with geometric characteristics of natural lightning discharge, a stochastic stepped leader propagation model was presented based on the fractal theory. By comparing simulation results with the statistics of natural lightning discharges, it was found that the fractal dimension of lightning trajectory in simulation was in the range of that observed in nature and the calculation results of electric field at ground level were in good agreement with the measurements of a negative flash, which shows the validity of this proposed model. Furthermore, a new equation to estimate the lightning striking distance to flat ground was suggested based on the present model. The striking distance obtained by this new equation is smaller than the value estimated by previous equations, which indicates that the traditional equations may somewhat overestimate the attractive effect of the ground.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Final Technical Report "Multiscale Simulation Algorithms for Biochemical Systems"

Biochemical systems are inherently multiscale and stochastic. In microscopic systems formed by living cells, the small numbers of reactant molecules can result in dynamical behavior that is discrete and stochastic rather than continuous and deterministic. An analysis tool that respects these dynamical characteristics is the stochastic simulation algorithm (SSA, Gillespie, 1976), a numerical simulation procedure that is essentially exact for chemical systems that are spatially homogeneous or well stirred. Despite recent improvements, as a procedure that simulates every reaction event, the SSA is necessarily inefficient for most realistic problems. There are two main reasons for this, both arising from the multiscale nature of the underlying problem: (1) stiffness, i.e. the presence of multiple timescales, the fastest of which are stable; and (2) the need to include in the simulation both species that are present in relatively small quantities and should be modeled by a discrete stochastic process, and species that are present in larger quantities and are more efficiently modeled by a deterministic differential equation (or at some scale in between). This project has focused on the development of fast and adaptive algorithms, and the fun- damental theory upon which they must be based, for the multiscale simulation of biochemical systems. Areas addressed by this project include: (1) Theoretical and practical foundations for ac- celerated discrete stochastic simulation (tau-leaping); (2) Dealing with stiffness (fast reactions) in an efficient and well-justified manner in discrete stochastic simulation; (3) Development of adaptive multiscale algorithms for spatially homogeneous discrete stochastic simulation; (4) Development of high-performance SSA algorithms.

97 MATHEMATICS AND COMPUTING↗

Evaluating Energy Impacts and Costs from PV Component Failures

The PV Reliability PerformanceModel (PV-RPM) tool is used to simulate the cost and energy impacts of component faults and failures for a hypothetical PV system. This model, originally developed by Sandia National Laboratories, is a new feature in the National Renewable Energy Laboratory's System Advisor Model (SAM), which performs stochastic analysissampling of probability distributions for faults, failures, and repairs. One PV system was created for this analysis to be representative of a portfolio of maintenance data maintained by Sandia National Laboratories. Fault,failure, and repair distributions from this system are varied from current baseline conditions to simulate different reliability and repair scenarios and resulting energy and cost impacts. Results highlight ways to express the uncertainty around PV system performance when considering the probabilistic behavior of components inthe system, and ways the PV-RPM model can be used to explore different failure and repair scenarios.

14 SOLAR ENERGY↗