Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “stochastic methods”

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 1,477 records · Page 82
All sourcesNASA NTRSDOE OSTI
DOE OSTI2013

A convergence study for SPDEs using combined Polynomial Chaos and Dynamically-Orthogonal schemes

We study the convergence properties of the recently developed Dynamically Orthogonal (DO) field equations [1] in comparison with the Polynomial Chaos (PC) method. To this end, we consider a series of one-dimensional prototype SPDEs, whose solution can be expressed analytically, and which are associated with both linear (advection equation) and nonlinear (Burgers equation) problems with excitations that lead to unimodal and strongly bi-modal distributions. We also propose a hybrid approach to tackle the singular limit of the DO equations for the case of deterministic initial conditions. The results reveal that the DO method converges exponentially fast with respect to the number of modes (for the problems considered) giving same levels of computational accuracy comparable with the PC method but (in many cases) with substantially smaller computational cost compared to stochastic collocation, especially when the involved parametric space is high-dimensional.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2011

Speckle evolution with multiple steps of least-squares phase removal

We study numerically the evolution of speckle fields due to the annihilation of optical vortices after the least-squares phase has been removed. A process with multiple steps of least-squares phase removal is carried out to minimize both vortex density and scintillation index. Statistical results show that almost all the optical vortices can be removed from a speckle field, which finally decays into a quasiplane wave after such an iterative process.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2025

Unsupervised anomaly clustering via offset alignment in multivariate grid sensing data

Modern industries increasingly rely on multi-sensor technologies to acquire complex, high-dimensional data streams, enabling advanced monitoring and control systems. One critical application is online anomaly detection in electrical smart grids, where multivariate and multimodal sensing technologies play a vital role. However, detecting anomalies in such time-series data is challenging due to their inherent temporal dependencies and stochastic behavior. Traditional approaches based on supervised and semi-supervised learning methods depend on labeled datasets, which are often unavailable in real-world scenarios. While unsupervised methods have emerged as promising alternatives, these methods are highly susceptible to noise and outliers commonly present in sensing applications. Furthermore, deep learning-based anomaly detection methods, despite their performance, are often criticized for their black-box nature, limiting their applicability in safety-critical and online environments where interpretability and explainability are paramount. In this work, we propose an unsupervised anomaly clustering method leveraging a cyclic alignment-based offset detection algorithm for multivariate time-series signals. The proposed method is applied to multivariate data collected from vibrational, voltage, and magnetic field sensors deployed in a local grid substation. Our results demonstrate the robustness of the algorithm in accurately clustering various anomalies/events across different sensing modalities. Additionally, we compare the effectiveness of the proposed approach against a simple pattern-based anomaly detection method, which performs well for univariate data but fails to generalize to multivariate and multimodal time-series data.

Mukherjee, Subrata [ORNL] (ORCID:0000000309930338)↗
DOE OSTI2024

A Distributionally Robust Optimization Framework for Stochastic Assessment of Power System Flexibility in Economic Dispatch

Given the complexity of power systems, particularly the high-dimensional variability of net loads, accurately depicting the entire operational range of net loads poses a challenge. To address this, recent methodologies have sought to gauge the maximum range of net load uncertainty across all buses. In this paper, we consider the stochastic nature of the net load and introduce a distributionally robust optimization framework that assesses system flexibility stochastically, accommodating a minimal extent of system violations. We verify the proposed method by solving the flexibility of the economic dispatch problem on four distinct IEEE standard test systems. Compared to traditional deterministic flexibility evaluations, our approach consistently yields less conservative flexibility outcomes.

distributionally robust optimization↗
DOE OSTI2018

Oscillations and Waves in a Nonlinear System with the 1/f Spectrum

Numerical methods are used to study a spatially distributed system of two nonlinear stochastic equations that simulate interacting phase transitions. Conditions for self-oscillations and waves are determined. The 1/f and 1/k spectra of extreme fluctuations are formed when waves emerge and move under the action of white noise. The distribution of the extreme fluctuations corresponds to the maximum entropy, which is proven by the stability of the 1/f and 1/k spectra. The formation and motion of waves under external periodic perturbation are accompanied by spatiotemporal chaotic resonance in which the domain of periodic pulsations is extended under the action of white noise.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2018

Mathematical Foundations for Uncertainty Quantification in Materials Design. Final Report

Final report DE-SC0010723. Our key accomplishment in the first year of the grant is that we started developing pathwise information theory-based and goal-oriented sensitivity analysis and parameter identification methods for complex high-dimensional dynamics and in particular of nonequilibrium extended systems. The combination of these novel methodologies provide the first methods in the literature which are capable to handle UQ questions for stochastic complex systems with some or all of the following features: (a) multi-scale models with a very large number of parameters, (b) spatially distributed systems such as Kinetic Monte Carlo or Langevin Dynamics, (c) non-equilibrium processes typically associated with coupled physico-chemical mechanisms, driven boundary conditions, etc. The first two such publications sponsored by the grant, have just been published.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Super Resolving Unrolled Neural Networks for Remote Sensing

In remote sensing systems, the capabilities of the system are constrained by the complex interactions between size, weight, and power (SWAP) of potential designs. In electro-optical (EO) systems, examples of these critical parameters include the system’s sensitivity and resolution. Those parameters can be increased by ever larger optical apertures and focal planes but at the cost of more SWAP. Multi-image super resolution (MISR) techniques allow resolution to be enhanced via computation rather than more sophisticated optical hardware. These algorithms combine multiple images together into a single, higher resolution image, trading temporal resolution and computation for spatial resolution. Fielded MISR techniques, such as Drizzle, can require several hundred images to create a single super resolved image, implying reduced temporal resolution, increased data acquisition load, and limiting mission applications. Iterative techniques, such as model-based image reconstruction and compressive sensing, have been shown to create super resolved images using fewer images than Drizzle. They do this by posing an optimization problem that balances accuracy between a highly accurate physical model and an image model. In the case of super resolution, the physical model is defined by the relation between low resolution input images and the desired high resolution output image. The image model encodes some assumptions about the super resolved image. These assumptions are meant to suppress reconstruction artifacts that arise due to deterministic physical model error, stochastic measurement noise, and potential undersampling. In practice, the performance of iterative methods are limited by imaging models compatible with optimization. Deep learning-based methods can effectively learn image models of arbitrary complexity, but lack the theoretical explainability and robustness of iterative techniques. Consensus equilibrium (CE) generalizes the iterative techniques beyond optimization, enabling blackbox algorithms such as traditional and neural image denoisers to be used as the image model. CE-based approaches retain much of the explainability and robustness of iterative techniques while allowing the expressiveness of machine learning image models to be used. Additionally, by unrolling iterations of CE with an embedded image denoiser, the image denoiser can be further trained and specialized to the specific application with potentially higher quality reconstructions. Under this project, we demonstrated the feasibility of training an unrolled neural network based upon CE. While we didn’t train one, we showed that the CE process is differentiable and its gradient can be tractably computed. We also explored the usage of a variants of CE akin to generative neural works. Most importantly, we applied the CE framework to a number of problems including non-blind deconvolution, upsampling, single-image super resolution, MISR, event-based sensing, and saturated deconvolution. Our MISR prototype creates high quality reconstructions with an order of magnitude fewer images than previous approaches and, critically, produces these reconstructions fast enough for practical usage.

47 OTHER INSTRUMENTATION↗
NASA NTRS1979

Molecular phylogenetic trees - On the validity of the Goodman-Moore augmentation algorithm

A response is made to the reply of Nei and Tateno (1979) to the letter of Holmquist (1978) supporting the validity of the augmentation algorithm of Moore (1977) in reconstructions of nucleotide substitutions by means of the maximum parsimony principle. It is argued that the overestimation of the augmented numbers of nucleotide substitutions (augmented distances) found by Tateno and Nei (1978) is due to an unrepresentative data sample and that it is only necessary that evolution be stochastically uniform in different regions of the phylogenetic network for the augmentation method to be useful. The importance of the average value of the true distance over all links is explained, and the relative variances of the true and augmented distances are calculated to be almost identical. The effects of topological changes in the phylogenetic tree on the augmented distance and the question of the correctness of ancestral sequences inferred by the method of parsimony are also clarified.

Holmquist, R.↗
NASA NTRS1985

Probabilistic Finite Element Development

The probabilistic finite element computer program known as Numerical Evaluation of Stochastic Structures Under Stress (NESSUS) is being developed for the analysis of critical structural components for reusable space propulsion systems. First year efforts involve the formulation of the probabilistic analysis strategy and the development of a probabilistic linear analysis code. The ultimate goal of the 3-year program is the development of a finite element code capable of performing nonlinear dynamic analysis of structures having stochastic material properties, geometry, and boundary conditions and subjected to random loading. Three levels of sophistication are envisioned for the stochastic description of the structural problem, namely: (1) homogeneous random variable for stiffness, mass, damping, and external loading; (2) stochastic characterization of variables at the element level, with specified interelement correlations; and (3) stochastic interpolation of variables within a finite element. Two alternative probabilistic analysis methods will be developed, allowing for all three levels of modeling sophistication.

Nagtegaal, J.↗
NASA NTRS1987

Proceedings of the 3rd Annual SCOLE Workshop

Topics addressed include: modeling and controlling the Spacecraft Control Laboratory Experiment (SCOLE) configurations; slewing maneuvers; mathematical models; vibration damping; gravitational effects; structural dynamics; finite element method; distributed parameter system; on-line pulse control; stability augmentation; and stochastic processes.

Taylor, Lawrence W., Jr.↗
NASA NTRS1987

Mixing, transport and combustion in sprays

The state-of-the-art in analytical methods for sprays and drop/turbulence interactions in combustion processes is presented with reference to dilute sprays, and dilute dispersed flows, that contain well-defined dispersed-phase elements in volume fractions lower than 1 percent. Attention is given to the near-injector dense spray region. The methods tested are those of locally homogeneous flow, deterministic separated flow, and stochastic separated flow; the first of these was ineffective due to its ignoring finite interphase transport rates, and the second failed by ignoring turbulent dispersion. Stochastic separated flow methodology yielded encouraging results through its treatment of both finite interphase transport rates and dispersed-phase/turbulence interactions.

Faeth, G. M.↗
NASA NTRS1989

Design Of Combined Stochastic Feedforward/Feedback Control

Methodology accommodates variety of control structures and design techniques. In methodology for combined stochastic feedforward/feedback control, main objectives of feedforward and feedback control laws seen clearly. Inclusion of error-integral feedback, dynamic compensation, rate-command control structure, and like integral element of methodology. Another advantage of methodology flexibility to develop variety of techniques for design of feedback control with arbitrary structures to obtain feedback controller: includes stochastic output feedback, multiconfiguration control, decentralized control, or frequency and classical control methods. Control modes of system include capture and tracking of localizer and glideslope, crab, decrab, and flare. By use of recommended incremental implementation, control laws simulated on digital computer and connected with nonlinear digital simulation of aircraft and its systems.

Halyo, Nesim↗
DOE OSTI2016

PROTOCOL TO EVALUATE THE MOISTURE DURABILITY OF ENERGY-EFFICIENT WALLS

Walls account for about 8% of the energy used in residential buildings. This energy penalty can be reduced with higher insulation levels and increased airtightness. However, these measures can compromise the moisture durability and long-term performance of wall assemblies because they can lead to lower moisture tolerance due to reduced drying potential. To avert these problems, a moisture durability protocol was developed to evaluate the probability that an energy-efficient wall design will experience mold growth. This protocol examines the effects of moisture sources in walls through a combination of simulations and lab experiments, uses the mold growth index as the moisture durability indicator, and is based on a probabilistic approach that utilizes stochastically varying input parameters. The simulation tools used include a new validated method for taking into account the effects of air leakage in wall assemblies This paper provides an overview of the developed protocol, discussion of the probabilistic simulation approach and describes results from the evaluation of two wall assemblies in Climate Zones 2, 4, and 6. The protocol will be used to supply builders with wall designs that are energy efficient, moisture durable and cost-effective.

durability↗
DOE OSTI2019

Mesoscale to Microscale Coupling for a Wind Ramp Case over Complex Terrain

Within the wind energy community, it has become increasingly important to simulate realistic weather events at high fidelity in order to improve wind turbine design, wind farm efficiency, and power forecasting. One of the major difficulties in simulating atmospheric events with numerical models at high fidelity is in accurately transitioning from the mesoscale to the microscale. The mesoscale flow, when input into the large-eddy simulation (LES) domain, will be free of resolved turbulence; needing sufficient time and space (known as fetch) for resolved-scale turbulence to develop. In order to reduce this fetch, several methods exist in which the temperature field, wind field, or both, are stochastically perturbed in order to accelerate the generation of turbulence. However, for cases involving complex terrain, it remains unclear if these perturbation processes are necessary, and if the interaction of the flow over the terrain, alone, is sufficient to generate adequate turbulence. In this study, a wind ramp event observed at the National Wind Technology Center (NWTC) outside of Boulder, CO is simulated first with a mesoscale model and then, using the simulated mesoscale solution as initial and boundary conditions, within a microscale LES solver. The mesoscale solution generated by the Weather Research and Forecasting (WRF) model accurately captures the wind ramp event with only a slight temporal shift and a small temperature bias when compared to the meteorological towers in the NWTC site. This information is then taken as the initial and boundary conditions within the Simulator for Wind Farm Applications (SOWFA) microscale LES model with varying terrain inputs. These inputs include the high-resolution (roughly 30 m grid spacing) terrain data from the Shuttle Radar Topography Mission (SRTM) (Figure 1a), as well as, three intermediate terrain resolution data sets produced by applying a uniform convolution window to the SRTM data with varying window sizes (Figure 1b-d). These are compared to the WRF terrain data interpolated to the high-resolution microscale grid (Figure 1e) in order to determine the impact of low-resolution (about 1 km grid spacing) terrain data on turbulence generation. The wind field is compared at several heights and distances along the dominant wind flow axis between each of the terrain cases. It is found that the development of turbulence across the domain is accelerated as terrain information increases in fidelity. Further, because this ramp event at the location of interest is largely influenced by the interaction of the short-wave trough with the terrain, as the topographic information decreases in fidelity the accuracy of the ramp event is decreased. To make generalizations and describe the benefits and shortcomings of this approach, several of the existing perturbation techniques are employed and compared to that of turbulence generation through topographic features.

17 WIND ENERGY↗
DOE OSTI2024

Integrating Multi-Source Data for Bi-Level Traffic Simulator Calibration: A Literature Review and Highway Case Study

Traffic simulation serves as a powerful tool for pre-evaluating policies and technologies. In this context, simulation-based Dynamic traffic assignment (DTA) models are capable of capturing traffic dynamics. They are well-known as critical tools in controlling and predicting traffic situations. The reliability of simulation results heavily depends on the calibration process. Most studies in the literature formulate and calibrate simulators based on a single source of collected data or multiple data sets with the same spatiotemporal characteristics. However, in practice, traffic data is collected by various tools with usually different spatial and temporal resolutions. This study introduces a novel approach to taking into account diverse input data from a variety of sources. An iterative bi-level solution is proposed. to equally treat traffic flow and speed data. The upper level solves flow calibration with the exact solution method, and the lower level calibrates the speed with the simultaneous perturbation stochastic approximation (SPSA) algorithm. Subsequently, the effectiveness of the proposed model is investigated using data from a six-mile section of Nashville's I-24 highway in Tennessee. The results demonstrate that our proposed model creates an effective feedback loop between the optimizer and the simulator for calibrating flow and speed to reduce the error between simulated and real data.

42 ENGINEERING↗
DOE OSTI2019

Uncertainty quantification methodology for hyperbolic systems with application to blood flow in arteries

We present a Stochastic Finite Volume - ADER (SFV-ADER) methodology for Uncertainty Quantification (UQ) in the general framework of systems of hyperbolic balance laws with uncertainty in parameters. The resulting method is weakly intrusive, meaning that deterministic solvers need minor modifications to include random contributions, and has no theoretical accuracy barrier. An illustration of the second-order version for the viscous Burgers equation with uncertain viscosity coefficient is first given in detail, under different deterministic initial conditions; the attainment of the theoretically expected convergence rate is demonstrated empirically; results and features of our scheme are discussed. We then extend the SFV-ADER method to a non-linear hyperbolic system with source terms that models one-dimensional blood flow in arteries, assuming uncertainties in a set of parameters of the problem. Results are then compared with measurements in a well-defined 1:1 experimental replica of the network of largest arteries in the human systemic circulation. For a 99%-confidence level, severe influence of parameter fluctuations is seen in pressure profiles, while secondary effects on flow rate profiles become visible, particularly in terminal branches. In conclusion, results suggest that the proposed methodology could successfully be applied to other physical problems involving hyperbolic balance laws.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2017

Hypothesis testing of scientific Monte Carlo calculations

The steadily increasing size of scientific Monte Carlo simulations and the desire for robust, correct, and reproducible results necessitates rigorous testing procedures for scientific simulations in order to detect numerical problems and programming bugs. However, the testing paradigms developed for deterministic algorithms have proven to be ill suited for stochastic algorithms. In this paper we demonstrate explicitly how the technique of statistical hypothesis testing, which is in wide use in other fields of science, can be used to devise automatic and reliable tests for Monte Carlo methods, and we show that these tests are able to detect some of the common problems encountered in stochastic scientific simulations. We argue that hypothesis testing should become part of the standard testing toolkit for scientific simulations.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗