Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “learned priors”

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 415 records · Page 23
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Atikokan Digital Twin: Machine learning in a biomass energy system

The Atikokan Generating Station, operated by Ontario Power Generation, has a 200 MW, biomass-fired tower boiler that operates on a dispatch schedule with a five-minute cycle. The boiler is generally operated in the range of 40–100 MW using two of five burner levels. In order to optimize boiler performance, we propose the implementation of a unique digital twin. Our digital twin abstraction couples Bayesian inference from science-based models and from observations (machine learning) with decision theory to predict operating-variable set points that optimize the physical asset (the boiler) in the presence of uncertainty (artificial intelligence). We focus this paper on the continuous Bayesian machine learning part of the Atikokan Digital Twin; we discuss decision theory in a companion paper. We identify and learn about 12 operational, model, and measured-output parameters and their uncertainties from high-fidelity, science-based simulations of the Atikokan boiler and from the observed measurements at the power plant. Since the goal of the Atikokan Digital Twin is to implement it online in real time, we require fast function evaluations for the quantities of interest extracted from the simulations in the Bayesian analysis. We use Gaussian process regression/interpolation to create accurate, robust surrogate models. We define the Bayesian priors and likelihood function and solve for the posterior distributions of the 12 parameters. Here we then propagate these distributions (i.e., parameters with uncertainty) into the predicted distributions of 790 quantities of interest to learn about the relative importance of various sources of error including experimental, model, and operating-parameter errors.

09 BIOMASS FUELS↗
DOE OSTI2025

Weak Form Scientific Machine Learning: Test Function Construction for System Identification

Weak form Scientific Machine Learning (WSciML) is a recently developed framework for data-driven modeling and scientific discovery. It leverages the weak form of equation error residuals to provide enhanced noise robustness in system identification via convolving model equations with test functions, reformulating the problem to avoid direct differentiation of data. The performance, however, relies on wisely choosing a set of compactly supported test functions. In this work, we mathematically motivate a novel data-driven method for constructing Single-scale-Local reference functions for creating the set of test functions. Our approach numerically approximates the integration error introduced by the quadrature and identifies the support size for which the error is minimal, without requiring access to the model parameter values. Through numerical experiments across various models, noise levels, and temporal resolutions, we demonstrate that the selected supports consistently align with regions of minimal parameter estimation error. We also compare the proposed method against the strategy for constructing Multi-scale-Global (and orthogonal) test functions introduced in our prior work, demonstrating the improved computational efficiency.

FOS: Computer and information sciences↗
NASA NTRS2007

X-31 Mishap: Lessons Learned

The experimental X-31 High Angle of Attack Research Aircraft crashed during a 1995 test mission flight conducted by NASA at Edwards Air Force Base, California. The pilot lost control of the airplane and was forced to eject, sustaining a permanent back injury that ended his flying career. Prior to this incident the airplane had a perfect record of several hundred non-eventful flights supported by an experienced team. During the subsequent investigation by a mishap committee it was discovered that a series of cascading events contributed to this accident. Some of the identified contributing factors that resulted in this mishap are common to aircraft design and to flight-test in general. The mistakes and the solutions are presented here so that the flight-test community may consider and learn from them. The primary cause of the crash was icing and, ultimately, a complete blockage of the pitot-static nose probe. The icing was caused by a freak weather phenomenon that was neither expected nor known to exist on the day of the mishap. The normal probe had been replaced with a special Kiel probe to allow total pressure measurements of up to 70 degrees angle of attack for flight-test purposes. The Kiel probe did not include a heater, because it was assumed that the airplane would not be flown in the clouds or in conditions conducive to icing. This assumption was later proven to be incorrect. The iced Kiel probe caused incorrect gain scheduling in the flight control system, resulting in an unstable aircraft. This failure was essentially undetected because of a faulty design in the flight control system architecture. There were, however, also a number of other issues that lead up to this situation that never should have happened. This presentation discusses what the issues were that contributed to the incident. After the incident was investigated, some of these issues were addressed and some changes were made. The second X-31 aircraft flew the remainder of the flight tests, and the program was successfully completed without incident. This presentation also shows a video of the mishap including lessons learned, and the changes that were made to resume the flight-test program are presented.

Larson, Richard R.↗
DOE OSTI2021

Consensus Equilibrium for Subsurface Delineation

Heterogeneity and insufficient site characterization limit our knowledge of the subsurface. Inversion techniques, which minimize the mismatch between observations and model predictions, have become an essential tool of subsurface characterization. Most optimization-based approaches fail to incorporate various implicit priors and capture the geological complexity. We overcome these limitations by deploying the plug-and-play and consensus equilibrium (CE) strategies, which provide a flexible framework for image reconstruction. Our CE methodology for spatial delineation of geologic formations consists of an image denoiser and a variational auto-encoder (deep learning-based emulator). The former ameliorates the reconstruction noise, yielding well-defined geological structures; its mathematical equivalence with the proximal operator allows the deployment of advanced denoisers (e.g., CNN-based denoiser) that do not correspond to a regularization objective. The latter defines a geology prior that imposes a geological constraint, for example, continuity and shape of geological features, onto the reconstructed image. Here, we conduct a series of numerical experiments dealing with transient two-dimensional flow driven by a pumping well and natural hydraulic head gradient. They demonstrate the CE framework's ability to delineate, both probabilistically and deterministically, complex subsurface environments with sufficient quality.

42 ENGINEERING↗
NASA NTRS2018

Know Your Enemy - Implementation of Bioremediation within a Suspected DNAPL Source Zone Following High-Resolution Site Characterization at Contractors Road Heavy Equipment Area, Kennedy Space Center, Florida

The National Aeronautics and Space Administration (NASA), Kennedy Space Center (KSC), has adopted high-resolution site characterization (HRSC) sampling techniques during baseline sampling prior to implementation of remedies to confirm and refine the conceptual site model (CSM). HRSC sampling was performed at Contractors Road Heavy Equipment Area (CRHE) prior to bioremediation implementation to verify the extent of the trichloroethene (TCE) dense non-aqueous phase liquid (DNAPL) source area (defined as the area with TCE concentrations above 1% solubility) and its daughter product dissolved plume that had been identified during previous HRSC events. The results of HRSC pre-bioremediation implementation sampling suggested that the TCE source area was larger than originally identified during initial site characterization activities, leading to a design refinement to improve electron donor distribution and increase the likelihood of achieving remedial objectives. Approach/Activities: HRSC was conducted from 2009 through 2014 to delineate the vertical and horizontal extent of chlorinated volatile organic compounds (CVOCs) in the groundwater. Approximately 2,340 samples were collected from 363 locations using direct push technology (DPT) groundwater sampling techniques. Samples were collected from up to 14 depth intervals at each location using a 4-foot sampling screen. This HRSC approach identified a narrow (approx. 5 to 30 feet wide), approximately 3,000 square foot TCE DNAPL source area (maximum detected TCE concentration of 160,000 micrograms per liter [micro-g/L] at DPT sampling location DPT0225). Prior to implementation of a bioremediation interim measure, HRSC baseline sampling was conducted using DPT groundwater sampling techniques. Concentrations of TCE were an order of magnitude lower than previous reported (12,000 micro-g/L maximum at DPT sampling location DPT0225) at locations sampled adjacent to previous sampling locations. To further evaluate the variability in concentrations observed additional sampling was conducted in 2016. The results identified higher concentrations than originally detected within the previously defined source area and the presence of source zone concentrations upgradient of the previously defined source area (maximum concentration observed 570,000 micro-g/L). The HRSC baseline sampling data allowed for a revision of the bioremediation design prior to implementation. Bioremediation was implemented within the eastern portion of the source area in November and December 2016 and quarterly performance monitoring was completed in March and June 2017. Reductions in CVOC concentrations from baseline were observed at all performance monitoring wells in the treatment area, and by June 2017, an approximate 95% CVOC mass reduction was observed based on monitoring well sampling results. Results/Lessons Learned: The results of this project suggest that, due to the complexity of DNAPL source zones, HRSC during pre-implementation baseline sampling in the TCE source zone was an essential strategy for verifying the treatment area and depth prior to remedy implementation. If the upgradient source zone mass was not identified prior to bioremediation implementation, the mass would have served as a long-term source for the dissolved plume.

Chrest, Anne↗
DOE OSTI2022

Physics-informed Bayesian machine learning case study: Integral blade rotors

This paper provides a physics-informed Bayesian machine learning (PIBML) description and case study. The PIBML approach applies three physics-based models to establish the initial beliefs before testing to determine the probability of milling stability (or prior). These include: receptance coupling substructure analysis (RCSA) prediction for the tool tip frequency response functions; finite element software prediction of the mechanistic force model coefficients; and a spindle speed-dependent power law model for process damping. Testing was then performed to identify optimal stable machining conditions using an expected improvement in material removal rate criterion. The prior probability of stability was updated using the test results to determine the posterior probability of stability. The test results were compared to the parameter recommendations provided by the endmill manufacturer. A demonstration integral blade rotor was machined at the optimal stable machining conditions for 304 stainless steel and 6061-T6 aluminum. Finally, the disagreement between manufacturer recommendations and milling performance in both materials tested emphasizes the need for broad implementation of PIBML approaches to increase machining productivity and efficiency.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
NASA NTRS1993

Population-based learning of load balancing policies for a distributed computer system

Effective load-balancing policies use dynamic resource information to schedule tasks in a distributed computer system. We present a novel method for automatically learning such policies. At each site in our system, we use a comparator neural network to predict the relative speedup of an incoming task using only the resource-utilization patterns obtained prior to the task's arrival. Outputs of these comparator networks are broadcast periodically over the distributed system, and the resource schedulers at each site use these values to determine the best site for executing an incoming task. The delays incurred in propagating workload information and tasks from one site to another, as well as the dynamic and unpredictable nature of workloads in multiprogrammed multiprocessors, may cause the workload pattern at the time of execution to differ from patterns prevailing at the times of load-index computation and decision making. Our load-balancing policy accommodates this uncertainty by using certain tunable parameters. We present a population-based machine-learning algorithm that adjusts these parameters in order to achieve high average speedups with respect to local execution. Our results show that our load-balancing policy, when combined with the comparator neural network for workload characterization, is effective in exploiting idle resources in a distributed computer system.

Mehra, Pankaj↗
DOE OSTI2023

Clustering High-dimensional Toxicogenomics Data with Rare Signals

Toxicogenomics studies the gene and protein activities to drug treatments or toxic exposures. As the drugs and genes are numerous, toxicogenomics data are naturally high dimensional, with dimension sizes up to millions. In addition, the distribution of toxicogenomics data is oftentimes skewed, and they contain rare but important signals representing a cell or organism’s response to toxicity. The combination of high dimension and extremely skewed distribution of toxicogenomics data makes clustering analysis extremely challenging.We present our study of clustering toxicogenomics data using classical approaches such as principal component analysis as well as deep learning approaches such as auto-encoders. Our experiments show that these approaches fail to preserve rare signals and produce high-quality clusters. We then explore augmenting matrix factorization with deep learning techniques such as attention mechanism to produce latent representations for clustering. Our technique is able to better preserve rare signals after dimensionality reduction than prior approaches. Furthermore, we combine our augmented matrix factorization with a mechanism similar to autoencoder to balance separable clusters and low regeneration errors. Our experiments demonstrate better clustering with our proposed approach.

Cong, Guojing↗
DOE OSTI2025

Conditional Pseudo-Reversible Normalizing Flow for Surrogate Modeling in Quantifying Uncertainty Propagation

We introduce a conditional pseudo-reversible normalizing flow (PR-NF) that directly learns conditional probability distributions from noisy physical models to efficiently quantify both forward and inverse uncertainty propagation. Traditional surrogate modeling approaches approximate only the deterministic component of physical models, requiring separate noise characterization and computationally expensive sampling methods for inverse problems. Here, in this work, we develop the conditional PR-NF model to directly learn and efficiently generate samples from the conditional probability density functions (PDFs). The training process utilizes dataset consisting of input-output pairs without requiring prior knowledge about the noise and the function. Once trained, our model efficiently generates samples from conditional PDFs for any input within the training domain. Moreover, the pseudo-reversibility feature allows for the use of fully connected neural network architectures, which simplifies the implementation and enables theoretical analysis. We provide a rigorous convergence analysis of the conditional PR-NF model, showing its ability to converge to the target conditional PDF using the Kullback−Leibler divergence. To demonstrate the effectiveness of our method, we apply it to several benchmark tests and a real-world geologic carbon storage problem.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1994

High performance compression of science data

Two papers make up the body of this report. One presents a single-pass adaptive vector quantization algorithm that learns a codebook of variable size and shape entries; the authors present experiments on a set of test images showing that with no training or prior knowledge of the data, for a given fidelity, the compression achieved typically equals or exceeds that of the JPEG standard. The second paper addresses motion compensation, one of the most effective techniques used in the interframe data compression. A parallel block-matching algorithm for estimating interframe displacement of blocks with minimum error is presented. The algorithm is designed for a simple parallel architecture to process video in real time.

Storer, James A.↗
NASA NTRS1994

High Performance Compression of Science Data

Two papers make up the body of this report. One presents a single-pass adaptive vector quantization algorithm that learns a codebook of variable size and shape entries; the authors present experiments on a set of test images showing that with no training or prior knowledge of the data, for a given fidelity, the compression achieved typically equals or exceeds that of the JPEG standard. The second paper addresses motion compensation, one of the most effective techniques used in interframe data compression. A parallel block-matching algorithm for estimating interframe displacement of blocks with minimum error is presented. The algorithm is designed for a simple parallel architecture to process video in real time.

Storer, James A.↗
NASA NTRS2004

System engineering for spaceborne optical interferometers

This paper is based on a collection of lessons-learned based on related missions including StarLight, a formation-flying stellar interferometry mission that merged with the Terrestrial Planet Finder technology development program just prior to entering phase C/D, the Space Interferometry Mission (currently in phase A), and the Shuttle Radar Topography Misssion of 2000.

system↗
DOE OSTI2020

Hydrodynamic and Radiographic Toolbox (HART)

With a multi-lab and university team, we propose to develop new methods for a Hydrodynamic and Radiographic Toolbox (HART) that will enable a fuller and more extensive use of experimental radiographic data towards better characterizing and reducing uncertainties in predictive modeling of weapons performance. We will achieve this by leveraging recent developments in areas of computational imaging, statistical and machine learning, and reduced order modeling of hydrodynamics. In terms of software practices, by partnering with XCP (RISTRA Project) we will conform to recent XCP standards that are inline with modern software practices and standards and ensure compatibility and ease of inter-operability with existing codes. Three new activities under this proposal include (a) the development and use of deep learning-based surrogates to accelerate reconstruction and variational inference of density fields from radiographs of hydrotests, (b) a model-data fusion strategy that couples deep learning-based density reconstructions with fast hydrodynamics simulators to better constrain the reconstruction, and (c) a method for treating asymmetries using techniques adopted from limited view tomography. All three new activities will be based on improved treatment of scatter, noise, beam spot movement, detector blur, and flat fielding in the forward model, and a use of sophisticated priors to aid in the re construction. The improvements to the forward model and improved algorithmic design of the reconstruction when complete will be contained in the iterative reconstruction code SHIVA—a code project that we have recently initiated. The many ways in which machine learning can be used in the reconstruction work will be contained in a code HERMES that has been initiated with DTRA support. For example, the significant levels of acceleration that will likely be achieved by the use of machine learning techniques will permit us (and are required) to quantify uncertainties in density retrievals. Next, the two-way coupling between density reconstruction and model-based simulation of the hydrodynamics will be contained in code EREBUS, and will permit a fuller realization of the potential of the data to constrain the hydrodynamic model and better address issues related to asymmetries in the problem. Finally, we anticipate that the better consistency with physics achieved in our reconstructions will allow them to be used by X-Division more so than today.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
DOE OSTI2020

Machine learning model to project the impact of COVID-19 on U.S. motor gasoline demand

Owing to the global lockdowns that resulted from the COVID-19 pandemic, fuel demand plummeted and the price of oil futures went negative in April 2020. Robust fuel demand projections are crucial to economic and energy planning and policy discussions. Here we incorporate pandemic projections and people's resulting travel and trip activities and fuel usage in a machine-learning-based model to project the US medium-term gasoline demand and study the impact of government intervention. We found that under the reference infection scenario, the US gasoline demand grows slowly after a quick rebound in May, and is unlikely to fully recover prior to October 2020. Under the reference and pessimistic scenario, continual lockdown (no reopening) could worsen the motor gasoline demand temporarily, but it helps the demand recover to a normal level quicker. Under the optimistic infection scenario, gasoline demand will recover close to the non-pandemic level by October 2020. The COVID 19 pandemic and consequent lockdown has had a substantial impact on mobility and therefore fuel demand and it is not clear when demand will recover. Ou et al. use a machine learning model that integrates health recovery scenarios to project the near-term future of gasoline demand.

COVID-19 pandemic↗
DOE OSTI2022

Hierarchical deep reinforcement learning reveals a modular mechanism of cell movement

Time-lapse images of cells and tissues contain rich information about dynamic cell behaviours, which reflect the underlying processes of proliferation, differentiation and morphogenesis. However, we lack computational tools for effective inference. Here we exploit deep reinforcement learning (DRL) to infer cell–cell interactions and collective cell behaviours in tissue morphogenesis from three-dimensional (3D) time-lapse images. We use hierarchical DRL (HDRL), known for multiscale learning and data efficiency, to examine cell migrations based on images with a ubiquitous nuclear label and simple rules formulated from empirical statistics of the images. When applied to Caenorhabditis elegans embryogenesis, HDRL reveals a multiphase, modular organization of cell movement. Imaging with additional cellular markers confirms the modular organization as a novel migration mechanism, which we term sequential rosettes. Furthermore, HDRL forms a transferable model that successfully differentiates sequential rosettes-based migration from others. Our study demonstrates a powerful approach to infer the underlying biology from time-lapse imaging without prior knowledge.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Optimizing Earthquake Nowcasting With Machine Learning: The Role of Strain Hardening in the Earthquake Cycle

Abstract Nowcasting is a term originating from economics, finance, and meteorology. It refers to the process of determining the uncertain state of the economy, markets or the weather at the current time by indirect means. In this paper, we describe a simple two‐parameter data analysis that reveals hidden order in otherwise seemingly chaotic earthquake seismicity. One of these parameters relates to a mechanism of seismic quiescence arising from the physics of strain‐hardening of the crust prior to major events. We observe an earthquake cycle associated with major earthquakes in California, similar to what has long been postulated. An estimate of the earthquake hazard revealed by this state variable time series can be optimized by the use of machine learning in the form of the Receiver Operating Characteristic skill score. The ROC skill is used here as a loss function in a supervised learning mode. Our analysis is conducted in the region of 5° × 5° in latitude‐longitude centered on Los Angeles, a region which we used in previous papers to build similar time series using more involved methods (Rundle & Donnellan, 2020, https://doi.org/10.1029/2020EA001097 ; Rundle, Donnellan et al., 2021, https://doi.org/10.1029/2021EA001757 ; Rundle, Stein et al., 2021, https://doi.org/10.1088/1361-6633/abf893 ). Here we show that not only does the state variable time series have forecast skill, the associated spatial probability densities have skill as well. In addition, use of the standard ROC and Precision (PPV) metrics allow probabilities of current earthquake hazard to be defined in a simple, straightforward, and rigorous way.

58 GEOSCIENCES↗
DOE OSTI2025

Evaluation of antifouling surfaces using a method that employs mussel larvae settlement quantified by machine learning

Antifouling coating development requires extensive performance testing. Coatings that prevent aquatic larval settlement are of interest because many forms of macrofouling begin at the larval stage. However, field testing can be time consuming and poorly controlled. Herein is reported a screening tool, Settlement of Larvae Assay using Mussels (SLAM), for down-selecting materials prior to field testing. The method entails using a dense concentration of mussel larvae that are allowed to settle on submerged test surfaces. Settled larvae are then quantified to provide a measure of antifouling performance. The SLAM test differentiated coatings with only slight differences in formulation. To enable efficient quantification of dense larvae settlement, an automated counting method was developed that combines two analyses: a color thresholding identifies larvae clumps, and a machine learning algorithm identifies non-clumped larvae. Finally, this automated ‘hybrid’ approach rapidly quantifies settled larvae as effectively as manual counting but in a fraction of the time.

Mytilus↗
DOE OSTI2025

Deep learning forecasts the spatiotemporal evolution of fluid-induced microearthquakes

Microearthquakes generated by subsurface fluid injection record the evolving stress state and permeability of reservoirs. Forecasting their spatiotemporal evolution is therefore critical for applications such as enhanced geothermal systems, carbon dioxide sequestration and other geoengineering applications. Here we propose a transformer neural network model that ingests hydraulic stimulation history and prior microearthquake observations to forecast four key quantities: cumulative microearthquake count, cumulative logarithmic seismic moment, and the 50th- and 95th-percentile extents of the microearthquake cloud. Applied to the EGS Collab Experiment 1 dataset, the model achieves R2 > 0.98 for the 1-s forecast horizon and R2 > 0.88 for the 15-s forecast horizon across all targets, and supplies uncertainty estimates through a learned standard deviation term. These accurate, uncertainty-quantified forecasts enable real-time inference of fracture propagation and permeability evolution, demonstrating the strong potential of deep-learning approaches to improve seismic-risk assessment and guide mitigation strategies in future fluid-injection operations.

Chung, Jaehong↗