Search NASA⌕ Search

SEARCH · Search NASA

Results for “Knowledge Modeling”

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 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

A Model Based Approach to Extract Health Information from Textual Data

In current nuclear power plants (NPPs) a large amount of condition-based data is being generated and stored to assess and monitor component health and performance. The format of this data can be either numeric (e.g., pump vibration data) or textual (e.g., condition report which assess component health). While assessing component health from numeric data can be performed with a large variety of methods, the extraction of information from textual data still remains a challenge. Natural language processing (NLP) methods are starting to be deployed in current NPPs mainly to filter out incident reports (IRs) that are not safety related by employing supervised machine learning methods. However, these methods do not really provide the quantitative information that might be contained in IRs. This paper presents an approach to extract information from textual data (e.g., from IRs, maintenance reports) that is based on NLP data analytics methods coupled with model-based system engineer (MBSE) models. NLP methods are employed to perform syntactic and semantic analyses. Syntactic analysis analyzes the grammatical structure of a sentence; such analysis includes: part of speech (POS) tagging (i.e., identification of grammatic elements of each string - e.g., nouns, verbs), named entity recognition (i.e., identification of text entities - e.g., names, dates, events), and relation extraction (e.g., coreference resolution). On the other hand, semantic analysis is designed to analyze the logic structure of a sentence. Through a specific set of rules, our methods can identify whether a sentence contains health information of a component (e.g., degraded performance, anomaly behavior) or the causal relationship between two events (i.e., a cause-effect pair). An innovative element of our approach is that semantic analysis relies on MBSE models to identify links between textual elements. MBSE are diagrams designed to represent system and component dependencies (from both a form and functional point of view). In our approach, MBSE models emulate system engineer knowledge about component/system architecture. This paper presents in detail how the integration of NLP methods and MBSE models is performed. Few analysis examples focusing on centrifugal pumps are presented.

97 - MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Improvement and generalization of ABCD method with Bayesian inference

To find New Physics or to refine our knowledge of the Standard Model at the LHC is an enterprise that involves many factors, such as the capabilities and the performance of the accelerator and detectors, the use and exploitation of the available information, the design of search strategies and observables, as well as the proposal of new models. We focus on the use of the information and pour our effort in re-thinking the usual data-driven ABCD method to improve it and to generalize it using Bayesian Machine Learning techniques and tools. We propose that a dataset consisting of a signal and many backgrounds is well described through a mixture model. Signal, backgrounds and their relative fractions in the sample can be well extracted by exploiting the prior knowledge and the dependence between the different observables at the event-by-event level with Bayesian tools. We show how, in contrast to the ABCD method, one can take advantage of understanding some properties of the different backgrounds and of having more than two independent observables to measure in each event. In addition, instead of regions defined through hard cuts, the Bayesian framework uses the information of continuous distribution to obtain soft-assignments of the events which are statistically more robust. To compare both methods we use a toy problem inspired by pp\to hh\to b\bar b b \bar b p p → h h → b b ‾ b b ‾ , selecting a reduced and simplified number of processes and analysing the flavor of the four jets and the invariant mass of the jet-pairs, modeled with simplified distributions. Taking advantage of all this information, and starting from a combination of biased and agnostic priors, leads us to a very good posterior once we use the Bayesian framework to exploit the data and the mutual information of the observables at the event-by-event level. We show how, in this simplified model, the Bayesian framework outperforms the ABCD method sensitivity in obtaining the signal fraction in scenarios with 1% and 0.5% true signal fractions in the dataset. We also show that the method is robust against the absence of signal. We discuss potential prospects for taking this Bayesian data-driven paradigm into more realistic scenarios.

Alvarez, Ezequiel↗
NASA NTRSDate not supplied

Uncertainty Quantification using Deep Ensembles for Decision Making in Cyber-Physical-Human Systems

In this paper and its companion, Differential Equation Approximation Using Gradient-Boosted Quantile Regression, Robison et al., we examine an approach to quantifying model uncertainty with the aim of increasing the trustworthiness of computational models in human-machine interactions. In Differential Equation Approximation Using Gradient-Boosted Quantile Regression, we focus on gradient-boosted decision trees, while in this one, we give more details about deep ensembles. Uncertainty quantification is crucial for building trustworthy autonomous decision-making agents in human-machine teams. There are two types of uncertainties: aleatoric and epistemic. The former is related to the inherent stochasticity (noise) of the process, whereas the latter is associated with the lack of knowledge or representation capability of models, such as neural networks. By lack of knowledge, we mean the model’s inability to accurately predict outputs for all possible inputs. The aleatory uncertainty can be estimated fairly easily with, for example, filters, whereas epistemic uncertainty is challenging to compute. This paper uses deep ensembles to quantify both aleatory and epistemic uncertainty. It can act as an uncertainty-aware surrogate transition model for decision-making frameworks. "Uncertainty-aware" means that the surrogate transition model should make predictions along with confidence in those predictions. In the context of decision-making, the transition models are ordinary differential equations (ODEs). Since ODEs can be simulated to make one-step or multi-step predictions, a good surrogate model for them should perform reasonably well in both modes. In a multi-step approach, the trajectory sampling method TS∞ was used to propagate uncertainty over multiple steps. The cartpole dynamical system was selected to demonstrate the ability of deep ensembles as good surrogate transition models for decision-making frameworks. The deep ensembles modeled the dynamics of cartpole ODEs and made uncertainty-aware predictions in single-step and multi-step transition modes.

CPH systems↗
NASA NTRS1991

Knowledge repositories for multiple uses

In the life cycle of a complex physical device or part, for example, the docking bay door of the Space Station, there are many uses for knowledge about the device or part. The same piece of knowledge might serve several uses. Given the quantity and complexity of the knowledge that must be stored, it is critical to maintain the knowledge in one repository, in one form. At the same time, because of quantity and complexity of knowledge that must be used in life cycle applications such as cost estimation, re-design, and diagnosis, it is critical to automate such knowledge uses. For each specific use, a knowledge base must be available and must be in a from that promotes the efficient performance of that knowledge base. However, without a single source knowledge repository, the cost of maintaining consistent knowledge between multiple knowledge bases increases dramatically; as facts and descriptions change, they must be updated in each individual knowledge base. A use-neutral representation of a hydraulic system for the F-111 aircraft was developed. The ability to derive portions of four different knowledge bases is demonstrated from this use-neutral representation: one knowledge base is for re-design of the device using a model-based reasoning problem solver; two knowledge bases, at different levels of abstraction, are for diagnosis using a model-based reasoning solver; and one knowledge base is for diagnosis using an associational reasoning problem solver. It was shown how updates issued against the single source use-neutral knowledge repository can be propagated to the underlying knowledge bases.

Williamson, Keith↗
NASA NTRS1988

AI and simulation: What can they learn from each other

Simulation and Artificial Intelligence share a fertile common ground both from a practical and from a conceptual point of view. Strengths and weaknesses of both Knowledge Based System and Modeling and Simulation are examined and three types of systems that combine the strengths of both technologies are discussed. These types of systems are a practical starting point, however, the real strengths of both technologies will be exploited only when they are combined in a common knowledge representation paradigm. From an even deeper conceptual point of view, one might even argue that the ability to reason from a set of facts (i.e., Expert System) is less representative of human reasoning than the ability to make a model of the world, change it as required, and derive conclusions about the expected behavior of world entities. This is a fundamental problem in AI, and Modeling Theory can contribute to its solution. The application of Knowledge Engineering technology to a Distributed Processing Network Simulator (DPNS) is discussed.

Colombano, Silvano P.↗
NASA NTRS1994

A toolbox and record for scientific models

Computational science presents a host of challenges for the field of knowledge-based software design. Scientific computation models are difficult to construct. Models constructed by one scientist are easily misapplied by other scientists to problems for which they are not well-suited. Finally, models constructed by one scientist are difficult for others to modify or extend to handle new types of problems. Construction of scientific models actually involves much more than the mechanics of building a single computational model. In the course of developing a model, a scientist will often test a candidate model against experimental data or against a priori expectations. Test results often lead to revisions of the model and a consequent need for additional testing. During a single model development session, a scientist typically examines a whole series of alternative models, each using different simplifying assumptions or modeling techniques. A useful scientific software design tool must support these aspects of the model development process as well. In particular, it should propose and carry out tests of candidate models. It should analyze test results and identify models and parts of models that must be changed. It should determine what types of changes can potentially cure a given negative test result. It should organize candidate models, test data, and test results into a coherent record of the development process. Finally, it should exploit the development record for two purposes: (1) automatically determining the applicability of a scientific model to a given problem; (2) supporting revision of a scientific model to handle a new type of problem. Existing knowledge-based software design tools must be extended in order to provide these facilities.

Ellman, Thomas↗
NASA NTRS1991

Outline for a theory of intelligence

Intelligence is defined as that which produces successful behavior. Intelligence is assumed to result from natural selection. A model is proposed that integrates knowledge from research in both natural and artificial systems. The model consists of a hierarchical system architecture wherein: (1) control bandwidth decreases about an order of magnitude at each higher level, (2) perceptual resolution of spatial and temporal patterns contracts about an order-of-magnitude at each higher level, (3) goals expand in scope and planning horizons expand in space and time about an order-of-magnitude at each higher level, and (4) models of the world and memories of events expand their range in space and time by about an order-of-magnitude at each higher level. At each level, functional modules perform behavior generation (task decomposition planning and execution), world modeling, sensory processing, and value judgment. Sensory feedback control loops are closed at every level.

Albus, James S.↗
NASA NTRS2003

A Numerical Climate Observing Network Design Study

This project was concerned with three related questions of an optimal design of a climate observing system: 1. The spatial sampling characteristics required from an ARGO system. 2. The degree to which surface observations from ARGO can be used to calibrate and test satellite remote sensing observations of sea surface salinity (SSS) as it is anticipated now. 3. The more general design of an climate observing system as it is required in the near future for CLIVAR in the Atlantic. An important question in implementing an observing system is that of the sampling density required to observe climate-related variations in the ocean. For that purpose this project was concerned with the sampling requirements for the ARGO float system, but investigated also other elements of a climate observing system. As part of this project we studied the horizontal and vertical sampling characteristics of a global ARGO system which is required to make it fully complementary to altimeter data with the goal to capture climate related variations on large spatial scales (less thanAttachment: 1000 km). We addressed this question in the framework of a numerical model study in the North Atlantic with an 1/6 horizontal resolution. The advantage of a numerical design study is the knowledge of the full model state. Sampled by a synthetic float array, model results will therefore allow to test and improve existing deployment strategies with the goal to make the system as optimal and cost-efficient as possible. Attachment: "Optimal observations for variational data assimilation".

Stammer, Detlef↗
NASA NTRS2011

Monte Carlo Simulations of the Formation Flying Dynamics for the Magnetospheric Multiscale (MMS) Mission

The MMS mission is an ambitious space physics mission that will fly 4 spacecraft in a tetrahedron formation in a series of highly elliptical orbits in order to study magnetic reconnection in the Earth's magnetosphere. The mission design is comprised of a combination of deterministic orbit adjust and random maintenance maneuvers distributed over the 2.5 year mission life. Formal verification of the requirements is achieved by analysis through the use of the End-to-End (ETE) code, which is a modular simulation of the maneuver operations over the entire mission duration. Error models for navigation accuracy (knowledge) and maneuver execution (control) are incorporated to realistically simulate the possible maneuver scenarios that might be realized These error models, coupled with the complex formation flying physics, lead to non-trivial effects that must be taken into account by the ETE automation. Using the ETE code, the MMS Flight Dynamics team was able to demonstrate that the current mission design satisfies the mission requirements.

Schiff, Conrad↗
NASA NTRS2023

Fault Detection and Diagnosis in Spacecraft Electrical Power Systems

The ability to accurately identify and isolate failures in the electrical power system (EPS) is critical to ensure the reliability of spacecraft. This paper proposes a novel solution to the problem of fault detection and diagnosis in direct current (DC) electric power systems for spacecraft. Autonomous operation becomes essential during deep space missions that lack the ability to monitor and control the spacecraft from ground locations. The current state of EPS fault supervision is insufficient to guarantee highly reliable operation. To solve this issue, a combination of model-based and knowledge-based techniques are used in a hierarchical framework to improve the diagnostic performance of the system. Noise, disturbances, and modeling errors are considered in the design of the fault detection system. Practical considerations related to spacecraft flight hardware and software are accounted for in the system design for flight applications. To assess the functionality of the design, a wide array of failures are simulated in a series of experiments. The experiments showed that the technique improved the capability of the autonomous system by increasing the number of fault types diagnosed. The significance of this study is to provide a framework capable of advanced diagnostics of an EPS with little to no interaction from human operators.

Autonomous Power Systems↗
DOE OSTI2025

RingX: Scalable Parallel Attention for Long-Context Learning on HPC

The attention mechanism has become foundational for remarkable AI breakthroughs since the introduction of the Transformer, driving the demand for increasingly longer context to power frontier models such as large-scale reasoning language models and high-resolution image/video generators. However, its quadratic computational and memory complexities present substantial challenges. Current state-of-the-art parallel attention methods, such as ring attention, are widely adopted for long-context training but utilize a point-to-point communication strategy that fails to fully exploit the capabilities of modern HPC network architectures. In this work, we propose ringX, a scalable family of parallel attention methods optimized explicitly for HPC systems. By enhancing workload partitioning, refining communication patterns, and improving load balancing, ringX achieves up to 3.4 × speedup compared to conventional ring attention on the Frontier supercomputer. Optimized for both bi-directional and causal attention mechanisms, ringX demonstrates its effectiveness through training benchmarks of a Vision Transformer (ViT) on a climate dataset and a Generative Pre-Trained Transformer (GPT) model, Llama3 8B. Our method attains an end-to-end training speedup of approximately 1.5 × in both scenarios. To our knowledge, the achieved 38% model FLOPs utilization (MFU) for training Llama3 8B with a 1M-token sequence length on 4,096 GPUs represents one of the highest training efficiencies reported for long-context learning on HPC systems. Our code implementation is available at https://github.com/jqyin/ringX-attention.

Yin, Junqi [ORNL] (ORCID:0000000338435520)↗
NASA NTRS2011

Decentralized Adaptive Control of Systems with Uncertain Interconnections, Plant-Model Mismatch and Actuator Failures

Decentralized adaptive control is considered for systems consisting of multiple interconnected subsystems. It is assumed that each subsystem s parameters are uncertain and the interconnection parameters are not known. In addition, mismatch can exist between each subsystem and its reference model. A strictly decentralized adaptive control scheme is developed, wherein each subsystem has access only to its own state but has the knowledge of all reference model states. The mismatch is estimated online for each subsystem and the mismatch estimates are used to adaptively modify the corresponding reference models. The adaptive control scheme is extended to the case with actuator failures in addition to mismatch.

Patre, Parag↗
NASA NTRS1988

Diurnal variability of the planetary albedo - An appraisal with satellite measurements and general circulation models

An atmospheric radiation model is used here to illustrate several features associated with modeling the diurnal cycle of the planetary albedo. It is found that even for clear regions there appear to be deficiencies in our knowledge of how to model this quantity. The diurnal amplitude factor, defined as the ratio of the diurnally averaged planetary albedo to that at noon, between two GCMs and measurements made from a geostationary satellite. While reasonable consistency is found, the comparisons underscore difficulties associated with converting local-time albedo measurements, as made from sun-synchronous satellites, to diurnally averaged albedos.

Potter, G. L.↗
NASA NTRS1983

A one-dimensional unsteady model of dual mode scramjet operation

A one-dimensional unsteady model and code have been developed which can handle internal flows involving area change, heat release, transients and coupled inviscid core-boundary layer flows. The model has been shown to work well in diverging channels with a supersonic entrance Mach number and unseparated boundary layers. Transient flows have also been studied. The model needs no prior knowledge of shock (or shocks) location and can be extended to two or three dimensions. The results show that the one-dimensional unsteady model developed here can be used for analysis of internal flows as occur in scramjet combustors.

Bussing, T. R. A.↗
NASA NTRS1991

Engineering technology for networks

Space Network (SN) modeling and evaluation are presented. The following tasks are included: Network Modeling (developing measures and metrics for SN, modeling of the Network Control Center (NCC), using knowledge acquired from the NCC to model the SNC, and modeling the SN); and Space Network Resource scheduling.

Paul, Arthur S.↗
NASA NTRS2011

Distributed Damage Estimation for Prognostics based on Structural Model Decomposition

Model-based prognostics approaches capture system knowledge in the form of physics-based models of components, and how they fail. These methods consist of a damage estimation phase, in which the health state of a component is estimated, and a prediction phase, in which the health state is projected forward in time to determine end of life. However, the damage estimation problem is often multi-dimensional and computationally intensive. We propose a model decomposition approach adapted from the diagnosis community, called possible conflicts, in order to both improve the computational efficiency of damage estimation, and formulate a damage estimation approach that is inherently distributed. Local state estimates are combined into a global state estimate from which prediction is performed. Using a centrifugal pump as a case study, we perform a number of simulation-based experiments to demonstrate the approach.

Daigle, Matthew↗
NASA NTRS2013

Automatic Parameter Tuning for the Morpheus Vehicle Using Particle Swarm Optimization

A high fidelity simulation using a PC based Trick framework has been developed for Johnson Space Center's Morpheus test bed flight vehicle. There is an iterative development loop of refining and testing the hardware, refining the software, comparing the software simulation to hardware performance and adjusting either or both the hardware and the simulation to extract the best performance from the hardware as well as the most realistic representation of the hardware from the software. A Particle Swarm Optimization (PSO) based technique has been developed that increases speed and accuracy of the iterative development cycle. Parameters in software can be automatically tuned to make the simulation match real world subsystem data from test flights. Special considerations for scale, linearity, discontinuities, can be all but ignored with this technique, allowing fast turnaround both for simulation tune up to match hardware changes as well as during the test and validation phase to help identify hardware issues. Software models with insufficient control authority to match hardware test data can be immediately identified and using this technique requires very little to no specialized knowledge of optimization, freeing model developers to concentrate on spacecraft engineering. Integration of the PSO into the Morpheus development cycle will be discussed as well as a case study highlighting the tool's effectiveness.

Birge, B.↗
NASA NTRSDate not supplied

Graph Convolutional Network-Strengthened Topic Modeling for Scientific Papers

Machine learning has been woven into statistics to modernize topic modeling over textual documents written in natural language, and scientific paper search and recommendation can consequently offer higher accuracy instead of counting on traditional keyword-based search. However, topic distribution of a paper resulted from existing topic modeling techniques only relies on the statistics of words contained in the paper itself. We argue that community users’ views of a paper may also provide insights at the time of recommendation. For example, if a paper on fake image detection has been cited heavily by machine learning papers, such a feature should be absorbed in the embedding of this paper, so that it can be recommended for future query on machine learning. In this paper, we present a Graph Convolutional Network-strengthened Topic Modeling (GCN-TM) method, which employs GCN technique to refine topic modeling of scientific papers. A citation-oriented knowledge graph is constructed, and topic modeling is mapped to feature embedding of the comprising papers. On top of its own topics carried in its content, each paper learns topics from its neighbors and revise its embedding accordingly. Our empirical studies over real-life scientific literature has proved the necessity and effectiveness of our proposed approach.

Jia Zhang↗